Skip to content

Visualization

batch_aabbs_to_mesh 

batch_aabbs_to_mesh(aabbs)

Convert a list of Open3D axis-aligned bounding boxes to a single triangle mesh.

Parameters:

  • aabbs (list of open3d.geometry.AxisAlignedBoundingBox) –

    List of bounding boxes to convert.

Returns:

  • TriangleMesh –

    Combined mesh of all bounding boxes.

Source code in occpy/visualization.py 
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
def batch_aabbs_to_mesh(aabbs):
    """
    Convert a list of Open3D axis-aligned bounding boxes to a single triangle mesh.

    Parameters
    ----------
    aabbs : list of open3d.geometry.AxisAlignedBoundingBox
        List of bounding boxes to convert.

    Returns
    -------
    open3d.geometry.TriangleMesh
        Combined mesh of all bounding boxes.
    """

    all_vertices = []
    all_triangles = []
    offset = 0

    F = np.array([
        [0, 1, 2], [0, 2, 3],       # bottom (z=min)
        [4, 5, 6], [4, 6, 7],       # top (z=max)
        [0, 1, 5], [0, 5, 4],       # y=min face
        [3, 2, 6], [3, 6, 7],       # y=max face
        [0, 3, 7], [0, 7, 4],       # x=min face
        [1, 2, 6], [1, 6, 5],       # x=max face
    ], dtype=np.int32)

    for aabb in aabbs:
        min_x, min_y, min_z = aabb.get_min_bound()
        max_x, max_y, max_z = aabb.get_max_bound()

        V = np.array([
            [min_x, min_y, min_z],
            [max_x, min_y, min_z],
            [max_x, max_y, min_z],
            [min_x, max_y, min_z],
            [min_x, min_y, max_z],
            [max_x, min_y, max_z],
            [max_x, max_y, max_z],
            [min_x, max_y, max_z],
        ], dtype=np.float64)

        all_vertices.append(V)
        all_triangles.append(F + offset)
        offset += V.shape[0]

    all_vertices = np.vstack(all_vertices)
    all_triangles = np.vstack(all_triangles)

    mesh = o3d.geometry.TriangleMesh()
    mesh.vertices = o3d.utility.Vector3dVector(all_vertices)
    mesh.triangles = o3d.utility.Vector3iVector(all_triangles)

    return mesh

darken_color 

darken_color(color, amount=0.6)

helper function to make provided color darker by amount

Parameters:

  • color –
  • amount –

Returns:

  • list of RGB colors (r, g, b) –
Source code in occpy/visualization.py 
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
def darken_color(color, amount=0.6):
    """
    helper function to make provided color darker by amount

    Parameters
    ----------
    color:
    amount

    Returns
    -------
    list of RGB colors (r, g, b)

    """
    r, g, b = to_rgb(color)
    return (r * amount, g * amount, b * amount)

get_Occl_TransectFigure 

get_Occl_TransectFigure(Nhit, Classification, OcclFrac, plot_dim, vox_dim, out_dir, start_ind=None, end_ind=None, axis=0, chm=None, vertBuffer=0, fig_prop=None, show_plots=False)

get_Occl_TransectFigure creates a matplotlib figure of a defined transect through the occlusion mapping output grid TODO: this function should be implemented in a more generic way!

Parameters:

  • Nhit –

    3D numpy array with number of hits in each grid cell (voxel)

  • Classification –

    3D numpy array with voxel Classification (Observed with hit = 1, Observed & empty = 2, Occluded = 3, Unobserved = 4)

  • OcclFrac –

    3D numpy array with Occlusion fraction

  • plot_dim –

    plot dimension of the input grid, as in [minX, minY, minZ, maxX, maxY, maxZ]

  • vox_dim –

    voxel dimensions in meters (cubic voxel are assumed)

  • out_dir –

    path to output directory

  • start_ind –

    voxel index of where the transect should start. If None [default] start_ind = 0

  • end_ind –

    voxel index of where the transect should end. If None [defaulte] end_ind = Nhit.shape[axis]

  • axis –

    axis index, either 0 (X-Axis), 1 (Y-Axis) or 2 (Z-Axis)

  • chm –

    2D canopy height model raster. if chm is None no CHM line will be plotted.

  • vertBuffer –

    optional vertical buffer added to the figure, if axis=0 or axis=1. This adds a padding above the canopy, so legend entries are not overlapping the actual transect.

  • fig_prop –

    python dictionary with figure properties. If fig_prop = None [default], the following settings will be defined: fig_prop = dict(fig_size=(3.14, 2.25), # figure size in inch lable_size=8, # font size for labels (e.g. x, y, z-axis labels= label_size_ticks=6, # font size for tick-labels label_size_tiny=4, # font size for other labels (e.g. legend labels) out_format='png') # output format of figure file

  • show_plots –

    Whether output figures should be shown [will pause the execution until figure is closed] or not.

Source code in occpy/visualization.py 
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
def get_Occl_TransectFigure(Nhit, Classification, OcclFrac, plot_dim, vox_dim, out_dir, start_ind=None, end_ind=None, axis=0, chm=None, vertBuffer=0, fig_prop=None, show_plots=False):
    """
    get_Occl_TransectFigure creates a matplotlib figure of a defined transect through the occlusion mapping output grid
    TODO: this function should be implemented in a more generic way!


    Parameters
    ----------
    Nhit: np.ndarray
        3D numpy array with number of hits in each grid cell (voxel)
    Classification: np.ndarray
        3D numpy array with voxel Classification (Observed with hit = 1, Observed & empty = 2, Occluded = 3, Unobserved = 4)
    OcclFrac: np.ndarray
        3D numpy array with Occlusion fraction
    plot_dim: np.ndarray
        plot dimension of the input grid, as in [minX, minY, minZ, maxX, maxY, maxZ]
    vox_dim: float
        voxel dimensions in meters (cubic voxel are assumed)
    out_dir: str
        path to output directory
    start_ind: int [default: None]
        voxel index of where the transect should start. If None [default] start_ind = 0
    end_ind: int [default: None]
        voxel index of where the transect should end. If None [defaulte] end_ind = Nhit.shape[axis]
    axis: int [0, 1, 2]
        axis index, either 0 (X-Axis), 1 (Y-Axis) or 2 (Z-Axis)
    chm: np.ndarray [default=None]
        2D canopy height model raster. if chm is None no CHM line will be plotted.
    vertBuffer: float [default=0]
        optional vertical buffer added to the figure, if axis=0 or axis=1. This adds a padding above the canopy, so legend
        entries are not overlapping the actual transect.
    fig_prop: dict [default=None]
        python dictionary with figure properties. If fig_prop = None [default], the following settings will be defined:
        fig_prop = dict(fig_size=(3.14, 2.25),  # figure size in inch
                        lable_size=8,           # font size for labels (e.g. x, y, z-axis labels=
                        label_size_ticks=6,     # font size for tick-labels
                        label_size_tiny=4,      # font size for other labels (e.g. legend labels)
                        out_format='png')       # output format of figure file

    show_plots: bool [default=False]
        Whether output figures should be shown [will pause the execution until figure is closed] or not.

    """

    if fig_prop is None:
        fig_prop = dict(fig_size=(3.14, 2.25),
                        label_size=8,
                        label_size_ticks=6,
                        label_size_tiny=4,
                        out_format='png', )


    if start_ind is None:
        start_ind = 0
    if end_ind is None:
        end_ind = Nhit.shape[axis]

    grid_dim = (int((plot_dim[3] - plot_dim[0]) / vox_dim), int((plot_dim[4] - plot_dim[1]) / vox_dim), int((plot_dim[5] - plot_dim[2]) / vox_dim))

    chm_slice_ref = None
    if axis == 0:  # get YZ, project axis X
        Nhit_Slice = np.sum(Nhit[start_ind:end_ind, :, :], axis=axis)
        OcclFrac_Slice = np.sum(Classification[start_ind:end_ind, :, :] == 3, axis=axis) / (
                    end_ind - start_ind)
        if chm is not None:
            # chm is [ny, nx] so to get YZ we project axis 1
            chm_slice_ref = np.max(chm[:, start_ind:end_ind], axis=1)
    elif axis == 1:  # get XZ, project axis Y
        Nhit_Slice = np.sum(Nhit[:, start_ind:end_ind, :], axis=axis)
        # OcclFrac_Slice = np.sum(Classification[:, start_ind:end_ind, :] == 3, axis=axis) / (
        #         end_ind - start_ind)
        OcclFrac = OcclFrac[:, start_ind:end_ind, :]
        mask = (OcclFrac >= 0.8)

        # sum only where mask is True
        sum_vals = np.sum(np.where(mask, OcclFrac, 0), axis=axis)

        # count matching values along the axis
        count_vals = np.sum(mask, axis=axis)

        # Safe division: avoid divide by zero and assign default where count == 0
        with np.errstate(divide='ignore', invalid='ignore'):
            OcclFrac_Slice = np.divide(sum_vals, count_vals)
            OcclFrac_Slice[count_vals == 0] = 0

        if chm is not None:
            # chm is [ny, nx] so to get XZ we project axis 0
            chm_slice_ref = np.max(chm[start_ind:end_ind, :], axis=0)
    else:  # get a slice of Z-Axis
        Nhit_Slice = np.sum(Nhit[:, :, start_ind:end_ind], axis=axis)
        OcclFrac_Slice = np.sum(Classification[:, :, start_ind:end_ind] == 3, axis=axis) / (
                end_ind - start_ind)

    #NHits_Slice_log = np.log10(Nhit_Slice, where=(Nhit_Slice != 0))

    # we need to rotate the slice for visualization purposes
    OcclFrac_Slice = np.rot90(OcclFrac_Slice)
    NHit_Slice = np.rot90(Nhit_Slice)

    fig = plt.figure(figsize=fig_prop['fig_size'])
    ax = fig.add_subplot(1, 1, 1)
    x_axis_vect = None
    if axis == 0:
        ax.set_xlabel(f"Y [m]", fontsize=fig_prop['label_size'])
        ax.set_ylabel(f"Height a.g. [m]", fontsize=fig_prop['label_size'])
        extent = [plot_dim[1]-plot_dim[1], plot_dim[4]-plot_dim[1], 0, OcclFrac_Slice.shape[0] * vox_dim]
        if vertBuffer != 0:
            extent_buf = extent.copy()
            extent_buf[3] = extent_buf[3] + vertBuffer
            ax.axis(extent_buf)
        else:
            ax.axis(extent)
        x_axis_vect = np.linspace(start=plot_dim[1]-plot_dim[1], stop=plot_dim[4]-plot_dim[1], num=grid_dim[1])

    elif axis == 1:
        ax.set_xlabel(f"X [m]", fontsize=fig_prop['label_size'])
        ax.set_ylabel(f"Height a.g. [m]", fontsize=fig_prop['label_size'])
        extent = [plot_dim[0]-plot_dim[0], plot_dim[3]-plot_dim[0], 0, OcclFrac_Slice.shape[0] * vox_dim]
        if vertBuffer!=0:
            extent_buf = extent.copy()
            extent_buf[3] = extent_buf[3] + vertBuffer
            ax.axis(extent_buf)
        else:
            ax.axis(extent)
        x_axis_vect = np.linspace(start=plot_dim[0]-plot_dim[0], stop=plot_dim[3]-plot_dim[0], num=grid_dim[0])
    else:
        ax.set_xlabel(f"X [m]", fontsize=fig_prop['label_size'])
        ax.set_ylabel(f"Y [m]", fontsize=fig_prop['label_size'])
        extent = [plot_dim[0]-plot_dim[0], plot_dim[3]-plot_dim[0], plot_dim[1]-plot_dim[1], plot_dim[4]-plot_dim[1]]
        ax.axis(extent)

    # define tick label size
    plt.yticks(fontsize=fig_prop['label_size_ticks'])
    plt.xticks(fontsize=fig_prop['label_size_ticks'])

    reds_cmap = plt.get_cmap(name='inferno_r')
    reds_cmap.set_under('k', alpha=0)
    grey_cmap = plt.get_cmap(name='Grays_r')
    grey_cmap.set_under('k', alpha=0)
    # plot raster data

    p50, p99 = np.percentile(OcclFrac_Slice*100, [50, 99])

    im1 = ax.imshow(NHit_Slice, cmap=grey_cmap, norm=LogNorm(vmin=1, vmax=np.amax(NHit_Slice)), interpolation='none',
                    extent=extent, alpha=1, aspect='auto')
    im2 = ax.imshow(OcclFrac_Slice * 100, cmap=reds_cmap, vmin=p50, vmax=p99, clim=[p50, p99], interpolation='none',
                    alpha=0.75, aspect='auto',
                    extent=extent)

    # Define equally spaced horizontal slots for two colorbars and one legend
    n_slots = 3
    slot_width = 0.28
    margin = 0.05
    gap = (1 - 2*margin - n_slots * slot_width) / (n_slots - 1)
    slot_height = 0.05
    y_pos_axes = 0.98


    if x_axis_vect is not None:
        if chm is not None:
            chm_ref_plot = ax.plot(x_axis_vect, chm_slice_ref, label="ULS CHM")
            # chm_comp_plot = ax.plot(x_axis_vect, chm_slice_comp, label="Comp CHM", linestyle='--') #TODO: implement that!

            legend_ax = ax.inset_axes([slot_width + gap + margin, y_pos_axes, slot_width, slot_height])
            legend_ax.axis("off")
            legend = legend_ax.legend(handles=[chm_ref_plot[0]], loc='center', frameon=True, ncol=1, fontsize=fig_prop['label_size_ticks'])
            legend.get_frame().set_alpha(1)


    # define colorbars with position and dimension
    start_pos = 0
    cax1 = ax.inset_axes([margin, y_pos_axes, slot_width, slot_height])
    cb1 = plt.colorbar(im1, cax=cax1, orientation='horizontal')
    cb1.ax.tick_params(labelsize=fig_prop['label_size_tiny'])
    cb1.set_label("Nr. Hits", size=fig_prop['label_size_ticks'])

    # Change ticks to actual values
    ticks = [10, 100, 1000]
    cb1.set_ticks(ticks)
    cb1.set_ticklabels([str(t) for t in ticks])


    # Second colorbar for Occlusion
    start_pos = 2 * (slot_width + gap)
    cax2 = ax.inset_axes([2 * (slot_width + gap) + margin, y_pos_axes, slot_width, slot_height])
    cb2 = plt.colorbar(im2, cax=cax2, orientation='horizontal')
    cb2.set_label("Occlusion [%]", size=fig_prop['label_size_ticks'])
    cb2.ax.tick_params(labelsize=fig_prop['label_size_tiny'])


    # tight layout
    plt.tight_layout()

    # save figure
    if axis == 0:
        plt.savefig(
            os.path.join(out_dir, f"Occlusion_Slice_YZ_{start_ind}_{end_ind}_voxels.{fig_prop['out_format']}"),
            dpi=300, format=fig_prop['out_format'])
    elif axis == 1:
        plt.savefig(
            os.path.join(out_dir, f"Occlusion_Slice_XZ_{start_ind}_{end_ind}_voxels.{fig_prop['out_format']}"),
            dpi=300, format=fig_prop['out_format'])
    else:
        plt.savefig(
            os.path.join(out_dir, f"Occlusion_Slice_XY_{start_ind}_{end_ind}_voxels.{fig_prop['out_format']}"),
            dpi=300, format=fig_prop['out_format'])

    if show_plots:
        plt.show(block=True)
    else:
        plt.close()

get_Occl_TransectFigure_BinaryOcclusion 

get_Occl_TransectFigure_BinaryOcclusion(Nhit, Classification, plot_dim, vox_dim, out_dir, start_ind=None, end_ind=None, axis=0, chm=None, vertBuffer=0, nhit_max=100000, nhit_min=1, fig_prop=None, show_plots=False)

get_Occl_TransectFigure_BinaryOcclusion creates a matplotlib figure of a defined transect through the occlusion mapping output grid TODO: this function should be implemented in a more generic way and potentially be integrated into get_Occl_TransectFigure()

Parameters:

  • Nhit –

    3D numpy array with number of hits in each grid cell (voxel)

  • Classification –

    3D numpy array with voxel Classification (Observed with hit = 1, Observed & empty = 2, Occluded = 3, Unobserved = 4)

  • plot_dim –

    plot dimension of the input grid, as in [minX, minY, minZ, maxX, maxY, maxZ]

  • vox_dim –

    voxel dimensions in meters (cubic voxel are assumed)

  • out_dir –

    path to output directory

  • start_ind –

    voxel index of where the transect should start. If None [default] start_ind = 0

  • end_ind –

    voxel index of where the transect should end. If None [defaulte] end_ind = Nhit.shape[axis]

  • axis –

    axis index, either 0 (X-Axis), 1 (Y-Axis) or 2 (Z-Axis)

  • chm –

    2D canopy height model raster. TODO: check and implement behavior if chm is not provided.

  • vertBuffer –

    optional vertical buffer added to the figure, if axis=0 or axis=1. This adds a padding above the canopy, so legend entries are not overlapping the actual transect.

  • fig_prop –

    python dictionary with figure properties. If fig_prop = None [default], the following settings will be defined: fig_prop = dict(fig_size=(3.14, 2.25), # figure size in inch lable_size=8, # font size for labels (e.g. x, y, z-axis labels= label_size_ticks=6, # font size for tick-labels label_size_tiny=4, # font size for other labels (e.g. legend labels) out_format='png') # output format of figure file

  • show_plots –

    Whether output figures should be shown [will pause the execution until figure is closed] or not.

Source code in occpy/visualization.py 
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
def get_Occl_TransectFigure_BinaryOcclusion(Nhit, Classification, plot_dim, vox_dim, out_dir, start_ind=None, end_ind=None, axis=0, chm=None, vertBuffer=0, nhit_max=100000, nhit_min=1, fig_prop=None, show_plots=False):
    """
        get_Occl_TransectFigure_BinaryOcclusion creates a matplotlib figure of a defined transect through the occlusion mapping output grid
        TODO: this function should be implemented in a more generic way and potentially be integrated into get_Occl_TransectFigure()


        Parameters
        ----------
        Nhit: np.ndarray
            3D numpy array with number of hits in each grid cell (voxel)
        Classification: np.ndarray
            3D numpy array with voxel Classification (Observed with hit = 1, Observed & empty = 2, Occluded = 3, Unobserved = 4)
        plot_dim: np.ndarray
            plot dimension of the input grid, as in [minX, minY, minZ, maxX, maxY, maxZ]
        vox_dim: float
            voxel dimensions in meters (cubic voxel are assumed)
        out_dir: str
            path to output directory
        start_ind: int [default: None]
            voxel index of where the transect should start. If None [default] start_ind = 0
        end_ind: int [default: None]
            voxel index of where the transect should end. If None [defaulte] end_ind = Nhit.shape[axis]
        axis: int [0, 1, 2]
            axis index, either 0 (X-Axis), 1 (Y-Axis) or 2 (Z-Axis)
        chm: np.ndarray [default=None]
            2D canopy height model raster. TODO: check and implement behavior if chm is not provided.
        vertBuffer: float [default=0]
            optional vertical buffer added to the figure, if axis=0 or axis=1. This adds a padding above the canopy, so legend
            entries are not overlapping the actual transect.
        fig_prop: dict [default=None]
            python dictionary with figure properties. If fig_prop = None [default], the following settings will be defined:
            fig_prop = dict(fig_size=(3.14, 2.25),  # figure size in inch
                            lable_size=8,           # font size for labels (e.g. x, y, z-axis labels=
                            label_size_ticks=6,     # font size for tick-labels
                            label_size_tiny=4,      # font size for other labels (e.g. legend labels)
                            out_format='png')       # output format of figure file
        show_plots: bool [default=False]
            Whether output figures should be shown [will pause the execution until figure is closed] or not.

        """
    if fig_prop is None:
        fig_prop = dict(fig_size=(3.14, 2.25),
                        label_size=8,
                        label_size_ticks=6,
                        label_size_tiny=4,
                        out_format='png', )

    if start_ind is None:
        start_ind = 0
    if end_ind is None:
        end_ind = Nhit.shape[axis]

    grid_dim = (int((plot_dim[3] - plot_dim[0]) / vox_dim), int((plot_dim[4] - plot_dim[1]) / vox_dim), int((plot_dim[5] - plot_dim[2]) / vox_dim))

    chm_slice_ref = None
    if axis == 0:  # get YZ, project axis X
        Nhit_Slice = np.sum(Nhit[start_ind:end_ind, :, :], axis=axis)
        OcclFrac_Slice = np.sum(Classification[start_ind:end_ind, :, :] == 3, axis=axis) / (
                    end_ind - start_ind)
        if chm is not None:
            # chm is [ny, nx] so to get YZ we project axis 1
            chm_slice_ref = np.max(chm[:, start_ind:end_ind], axis=1)
    elif axis == 1:  # get XZ, project axis Y
        Nhit_Slice = np.sum(Nhit[:, start_ind:end_ind, :], axis=axis)
        OcclFrac_Slice = np.sum(Classification[:, start_ind:end_ind, :] == 3, axis=axis) / (end_ind - start_ind)
        if chm is not None:
            # chm is [ny, nx] so to get XZ we project axis 0
            chm_slice_ref = np.max(chm[start_ind:end_ind, :], axis=0)
    else:  # get a slice of Z-Axis
        Nhit_Slice = np.sum(Nhit[:, :, start_ind:end_ind], axis=axis)
        OcclFrac_Slice = np.sum(Classification[:, :, start_ind:end_ind] == 3, axis=axis) / (
                end_ind - start_ind)

    #NHits_Slice_log = np.log10(Nhit_Slice, where=(Nhit_Slice != 0))

    # we need to rotate the slice for visualization purposes
    OcclFrac_Slice = np.rot90(OcclFrac_Slice)
    NHit_Slice = np.rot90(Nhit_Slice)

    fig = plt.figure(figsize=fig_prop['fig_size'])
    ax = fig.add_subplot(1, 1, 1)
    x_axis_vect = None
    if axis == 0:
        ax.set_xlabel(f"Y [m]", fontsize=fig_prop['label_size'])
        ax.set_ylabel(f"Height a.g. [m]", fontsize=fig_prop['label_size'])
        extent = [plot_dim[1]-plot_dim[1], plot_dim[4]-plot_dim[1], 0, OcclFrac_Slice.shape[0] * vox_dim]
        if vertBuffer != 0:
            extent_buf = extent.copy()
            extent_buf[3] = extent_buf[3] + vertBuffer
            ax.axis(extent_buf)
        else:
            ax.axis(extent)
        x_axis_vect = np.linspace(start=plot_dim[1]-plot_dim[1], stop=plot_dim[4]-plot_dim[1], num=grid_dim[1])

    elif axis == 1:
        ax.set_xlabel(f"X [m]", fontsize=fig_prop['label_size'])
        ax.set_ylabel(f"Height a.g. [m]", fontsize=fig_prop['label_size'])
        extent = [plot_dim[0]-plot_dim[0], plot_dim[3]-plot_dim[0], 0, OcclFrac_Slice.shape[0] * vox_dim]
        if vertBuffer!=0:
            extent_buf = extent.copy()
            extent_buf[3] = extent_buf[3] + vertBuffer
            ax.axis(extent_buf)
        else:
            ax.axis(extent)
        x_axis_vect = np.linspace(start=plot_dim[0]-plot_dim[0], stop=plot_dim[3]-plot_dim[0], num=grid_dim[0])
    else:
        ax.set_xlabel(f"X [m]", fontsize=fig_prop['label_size'])
        ax.set_ylabel(f"Y [m]", fontsize=fig_prop['label_size'])
        extent = [plot_dim[0]-plot_dim[0], plot_dim[3]-plot_dim[0], plot_dim[1]-plot_dim[1], plot_dim[4]-plot_dim[1]]
        ax.axis(extent)

    # define tick label size
    plt.yticks(fontsize=fig_prop['label_size_ticks'])
    plt.xticks(fontsize=fig_prop['label_size_ticks'])

    reds_cmap = plt.get_cmap(name='plasma_r')
    reds_cmap.set_under('k', alpha=0)
    grey_cmap = plt.get_cmap(name='Grays_r')
    grey_cmap.set_under('k', alpha=0)
    # plot raster data


    im1 = ax.imshow(NHit_Slice, cmap=grey_cmap, norm=LogNorm(vmin=nhit_min, vmax=nhit_max), interpolation='none',
                    extent=extent, alpha=1, aspect='auto')
    im2 = ax.imshow(OcclFrac_Slice * 100, cmap=reds_cmap, vmin=1, vmax=50, clim=[1, 50], interpolation='none',
                    alpha=0.75, aspect='auto',
                    extent=extent)

    # Define equally spaced horizontal slots for two colorbars and one legend
    n_slots = 3
    slot_width = 0.28
    margin = 0.05
    gap = (1 - 2*margin - n_slots * slot_width) / (n_slots - 1)
    slot_height = 0.05
    y_pos_axes = 0.98


    if x_axis_vect is not None:
        if chm is not None:
            chm_ref_plot = ax.plot(x_axis_vect, chm_slice_ref, label="ULS CHM")
            # chm_comp_plot = ax.plot(x_axis_vect, chm_slice_comp, label="Comp CHM", linestyle='--') #TODO: implement that!

            legend_ax = ax.inset_axes([slot_width + gap + margin, y_pos_axes, slot_width, slot_height])
            legend_ax.axis("off")
            legend = legend_ax.legend(handles=[chm_ref_plot[0]], loc='center', frameon=True, ncol=1, fontsize=fig_prop['label_size_ticks'])
            legend.get_frame().set_alpha(1)


    # define colorbars with position and dimension
    start_pos = 0
    cax1 = ax.inset_axes([margin, y_pos_axes, slot_width, slot_height])
    cb1 = plt.colorbar(im1, cax=cax1, orientation='horizontal')
    cb1.ax.tick_params(labelsize=fig_prop['label_size_tiny'])
    cb1.set_label("Nr. Hits", size=fig_prop['label_size_ticks'])

    # Change ticks to actual values
    ticks = [10, 100, 1000]
    cb1.set_ticks(ticks)
    cb1.set_ticklabels([str(t) for t in ticks])


    # Second colorbar for Occlusion
    start_pos = 2 * (slot_width + gap)
    cax2 = ax.inset_axes([2 * (slot_width + gap) + margin, y_pos_axes, slot_width, slot_height])
    cb2 = plt.colorbar(im2, cax=cax2, orientation='horizontal')
    cb2.set_label("Occlusion [%]", size=fig_prop['label_size_ticks'])
    cb2.ax.tick_params(labelsize=fig_prop['label_size_tiny'])

    # tight layout
    plt.tight_layout()

    # save figure
    if axis == 0:
        plt.savefig(
            os.path.join(out_dir, f"Occlusion_Slice_YZ_{start_ind}_{end_ind}_voxels_binary.{fig_prop['out_format']}"),
            dpi=300, format=fig_prop['out_format'])
    elif axis == 1:
        plt.savefig(
            os.path.join(out_dir, f"Occlusion_Slice_XZ_{start_ind}_{end_ind}_voxels_binary.{fig_prop['out_format']}"),
            dpi=300, format=fig_prop['out_format'])
    else:
        plt.savefig(
            os.path.join(out_dir, f"Occlusion_Slice_XY_{start_ind}_{end_ind}_voxels_binary.{fig_prop['out_format']}"),
            dpi=300, format=fig_prop['out_format'])

    if show_plots:
        plt.show(block=True)
    else:
        plt.close()

get_Occlusion_ProfileFigure 

get_Occlusion_ProfileFigure(Classification, plot_dim, vox_dim, out_dir, low_thresh=0, vertBuffer=0, max_percentage=100, fig_prop=None, show_plots=False)

get_Occlusion_ProfileFigure produces a profile figure of Occluded, filled, and empty voxels

Parameters:

  • Classification –

    3D voxel grid with voxel classification (1=Observed with hit, 2 = observed empty, 3 = occlusion, 4 = unobserved)

  • plot_dim –

    plot dimension of the input grid, as in [minX, minY, minZ, maxX, maxY, maxZ]

  • vox_dim –

    voxel dimension in meters (assuming cubic voxels)

  • out_dir –

    directory for figure output

  • low_thresh –

    cut-off for lower part of the grid to exclude e.g. high occlusion towards the ground

  • vertBuffer –

    vertical buffer to add ontop of highest canopy point. e.g. needed to align Y-Axis with transect figure made with get_Occl_TransectFigure

  • max_percentage –

    maximum volume percentage to be shown on x-Axis

  • fig_prop – 
    python dictionary with figure properties. If fig_prop = None [default], the following settings will be defined:
fig_prop = dict(fig_size=(3.14, 2.25),  # figure size in inch
                lable_size=8,           # font size for labels (e.g. x, y, z-axis labels=
                label_size_ticks=6,     # font size for tick-labels
                label_size_tiny=4,      # font size for other labels (e.g. legend labels)
                out_format='png')       # output format of figure file
  • show_plots – 
    Whether output figures should be shown [will pause the execution until figure is closed] or not.
Source code in occpy/visualization.py 
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
def get_Occlusion_ProfileFigure(Classification, plot_dim, vox_dim, out_dir, low_thresh=0, vertBuffer=0, max_percentage=100, fig_prop=None, show_plots=False):
    """get_Occlusion_ProfileFigure produces a profile figure of Occluded, filled, and empty voxels

    Parameters
    ----------
    Classification: np.ndarray
        3D voxel grid with voxel classification (1=Observed with hit, 2 = observed empty, 3 = occlusion, 4 = unobserved)
    plot_dim: np.ndarray
        plot dimension of the input grid, as in [minX, minY, minZ, maxX, maxY, maxZ]
    vox_dim: float
        voxel dimension in meters (assuming cubic voxels)
    out_dir: str
        directory for figure output
    low_thresh: float, default 0
        cut-off for lower part of the grid to exclude e.g. high occlusion towards the ground
    vertBuffer: float, default 0
        vertical buffer to add ontop of highest canopy point. e.g. needed to align Y-Axis with transect figure made with
        get_Occl_TransectFigure
    max_percentage: float, default 100
        maximum volume percentage to be shown on x-Axis
    fig_prop: dict, default None
            python dictionary with figure properties. If fig_prop = None [default], the following settings will be defined:
            fig_prop = dict(fig_size=(3.14, 2.25),  # figure size in inch
                            lable_size=8,           # font size for labels (e.g. x, y, z-axis labels=
                            label_size_ticks=6,     # font size for tick-labels
                            label_size_tiny=4,      # font size for other labels (e.g. legend labels)
                            out_format='png')       # output format of figure file
    show_plots: bool, default False
            Whether output figures should be shown [will pause the execution until figure is closed] or not.
    """

    grid_dim = (int((plot_dim[3] - plot_dim[0]) / vox_dim), int((plot_dim[4] - plot_dim[1]) / vox_dim),
                int((plot_dim[5] - plot_dim[2]) / vox_dim))

    vert_vect = np.arange(start=low_thresh, stop=Classification.shape[2] * vox_dim, step=vox_dim)
    Classification = Classification[:,:,int(low_thresh / vox_dim):]
    # a hack to make sure that vert_vect is of the same length as OcclVertProf TODO: this has to be checked if it is generic!


    OcclVertProf = np.sum(Classification == 3, axis=0)
    OcclVertProf = np.sum(OcclVertProf, axis=0)
    OcclVertProf_Rel = OcclVertProf / ((grid_dim[0]) * (grid_dim[1]))

    FilledVertProf = np.sum(Classification == 1, axis=0)
    FilledVertProf = np.sum(FilledVertProf, axis=0)
    FilledVertProf_Rel = FilledVertProf / (grid_dim[0] * grid_dim[1])

    EmptyVertProf = np.sum(np.logical_or(Classification == 2, Classification==0), axis=0)
    EmptyVertProf = np.sum(EmptyVertProf, axis=0)
    EmptyVertProf_Rel = EmptyVertProf / (grid_dim[0] * grid_dim[1])

    heights = vert_vect[0:len(OcclVertProf)]

    percentages = np.column_stack([FilledVertProf_Rel*100, OcclVertProf_Rel*100, EmptyVertProf_Rel*100])
    categories = ['Filled', 'Occluded', 'Empty']
    colors = ['skyblue', 'salmon', 'lightgreen']

    # Compute cumulative percentages for stacking
    cumulative = np.cumsum(percentages, axis=1)

    palette = sns.color_palette('colorblind', n_colors=len(categories))
    palette[2] = (1.0, 1.0, 1.0) # white for empty

    fig, ax = plt.subplots(figsize=fig_prop['fig_size'])

    for i, cat in enumerate(categories):
        left = cumulative[:, i - 1] if i > 0 else np.zeros_like(heights)
        face_color = palette[i]
        edge_color = darken_color(face_color, 0.8)  # slightly darker for lines

        # Fill area
        ax.fill_betweenx(
            heights, left, cumulative[:, i],
            color=face_color, alpha=0.6
        )
        # Outline
        ax.plot(cumulative[:, i], heights, color=edge_color, linewidth=1.5, label="_nolegend_")

    ax.set_xlabel('Percentage of voxels [%]', fontsize=fig_prop['label_size'])
    ax.set_ylabel('Height above ground [m]', fontsize=fig_prop['label_size'])
    ax.set_xlim(0.1,max_percentage)
    ax.set_ylim(0,np.max(heights) + vertBuffer)
    plt.xticks(fontsize=fig_prop['label_size_ticks'])
    plt.yticks(fontsize=fig_prop['label_size_ticks'])
    ax.legend(categories[0:2], fontsize=fig_prop['label_size_ticks'])
    plt.tight_layout()

    plt.savefig(os.path.join(out_dir, f"OcclusionVertProf.{fig_prop['out_format']}"), dpi=300, format=fig_prop['out_format'])
    if show_plots:
        plt.show(block=True)
    else:
        plt.close()

interactive_figure 

interactive_figure(output_dir, axis=0)

Create an interactive slice viewer for voxel grids with occlusion overlay.

Loads Classification.npy and Nhit.npy, builds interactive figure with sliders to select the slice center and projection depth, and visualizes: - log10(Nhit) as a grayscale heatmap, and - fraction of voxels classified as occluded as a colored overlay.

Axis selects the slicing plane: - 0: YZ slice - 1: XZ slice - 2: XY slice

Parameters:

  • output_dir (str) –

    Directory containing Classification.npy and Nhit.npy arrays.

  • axis (int, default: 0 ) –

    Axis orthogonal to the slicing plane (0, 1, or 2).

Source code in occpy/visualization.py 
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
def interactive_figure(output_dir, axis=0):
    """
    Create an interactive slice viewer for voxel grids with occlusion overlay.

    Loads Classification.npy and Nhit.npy, builds interactive
    figure with sliders to select the slice center and projection depth, and
    visualizes:
      - log10(Nhit) as a grayscale heatmap, and
      - fraction of voxels classified as occluded as a colored overlay.

    Axis selects the slicing plane:
      - 0: YZ slice 
      - 1: XZ slice
      - 2: XY slice

    Parameters
    ----------
    output_dir : str
        Directory containing Classification.npy and Nhit.npy arrays.
    axis : int, default 0
        Axis orthogonal to the slicing plane (0, 1, or 2).
    """

    raise ValueError("Not working yet. Use get_Occl_TransectFigure instead for now.")

    classification_arr = np.load(os.path.join(output_dir, "Classification.npy"))
    nhit_arr = np.load(os.path.join(output_dir, "Nhit.npy"))

    print("Shapes: (X,Y,Z)")
    print(f"Classification: {classification_arr.shape}")
    print(f"NHIT: {nhit_arr.shape}")

    # -- function to generate plot for given parameters

    def generate_image(center, depth, axis):
        start_ind = int(max(0, center-depth/2))
        max_ind = classification_arr.shape[axis] - 1 
        end_ind = int(min(max_ind, start_ind+depth))

        if axis==0: # get a slice of X-Axis, YZ image
            Nhit_Slice = np.sum(nhit_arr[start_ind:end_ind, :, :], axis=axis)
            OcclFrac_Slice = np.sum(classification_arr[start_ind:end_ind, :, :]==3, axis=axis) / (end_ind - start_ind)
        elif axis==1: # XZ image
            Nhit_Slice = np.sum(nhit_arr[:,start_ind:end_ind,:], axis=axis)
            OcclFrac_Slice = np.sum(classification_arr[:, start_ind:end_ind, :] == 3, axis=axis) / (end_ind - start_ind)
        else: # XY image
            Nhit_Slice = np.sum(nhit_arr[:, :, start_ind:end_ind], axis=axis)
            OcclFrac_Slice = np.sum(classification_arr[:, :, start_ind:end_ind] == 3, axis=axis) / (end_ind - start_ind)

        NHits_Slice_log = np.log10(Nhit_Slice, where=(Nhit_Slice != 0))
        OcclFrac_Slice = np.rot90(OcclFrac_Slice)
        NHits_Slice_log = np.rot90(NHits_Slice_log)
        vlim_occl = np.ceil(np.amax(OcclFrac_Slice*100) / 10.0) * 10
        MAX_value_nhits = np.amax(NHits_Slice_log)

        return NHits_Slice_log, OcclFrac_Slice, vlim_occl, MAX_value_nhits

    # -- init

    init_center = int(round(classification_arr.shape[axis]/2))
    init_depth = 50

    # ---- define figure thingies

    fig = plt.figure(figsize=(6, 4))
    # fig = plt.figure(figsize=(4, 8))
    # figure properties
    ax = fig.add_subplot(1,1,1)
    ax.set_aspect('equal', adjustable='box')
    # adjust the main plot to make room for the sliders
    # fig.subplots_adjust(left=0.1)
    ax.tick_params(axis='both', which='major', labelsize=11)
    ax.xaxis.label.set_size(12)
    ax.yaxis.label.set_size(12)
    ax.spines[['right', 'top']].set_visible(False)
    ax.spines[['left', 'bottom']].set_linewidth(.8)
    ax.spines[['left', 'bottom']].set_color('k')
    reds_cmap = plt.get_cmap(name='viridis_r')
    reds_cmap.set_under('k', alpha=0)
    greens_cmap = plt.get_cmap(name='Greys_r')
    greens_cmap.set_under('k', alpha=0)

    # --- define extent
    if axis==0:
        ax.set_xlabel(f"Y (m)")
        ax.set_ylabel(f"Z (m)")
        extent = [0, classification_arr.shape[1]*VOX_DIM, 0, classification_arr.shape[2]*VOX_DIM]
        ax.axis(extent)
    elif axis==1:
        ax.set_xlabel(f"X (m)")
        ax.set_ylabel(f"Z (m)")
        extent = [0, classification_arr.shape[0]*VOX_DIM, 0, classification_arr.shape[2]*VOX_DIM]
        ax.axis(extent)
    else:
        ax.set_xlabel(f"X (m)")
        ax.set_ylabel(f"Y (m)")
        extent = [0, classification_arr.shape[0]*VOX_DIM, 0, classification_arr.shape[1]*VOX_DIM]
        ax.axis(extent)

    # --- generate initial image and show

    nhit_img, occl_img, lim_ocll, lim_nhit = generate_image(init_center, init_depth, axis=axis)
    im1 = ax.imshow(nhit_img, cmap=greens_cmap, clim=[0.1, lim_nhit], interpolation='none',
                    alpha=1, extent=extent)
    im2 = ax.imshow(occl_img * 100, cmap=reds_cmap, clim=[1, lim_ocll], interpolation='none',
                    alpha=1, extent=extent)

    # --- add colorbars
    # TODO: update these based on the slice?
    axins1 = inset_axes(
        ax,
        width="35%",
        height="5%",
        loc="upper right",
    )
    axins1.xaxis.set_ticks_position("bottom")
    fig.colorbar(im1, cax=axins1, orientation='horizontal', label="Log Nr.Hits")
    axins2 = inset_axes(
        ax,
        width="35%",
        height="5%",
        loc="upper left",
    )
    axins2.xaxis.set_ticks_position("bottom")
    fig.colorbar(im2, cax=axins2, orientation='horizontal', label="Occluded voxels (%)")

    # --- add sliders for center and depth

    ax_center = fig.add_axes([0.15, 0.1, 0.0225, 0.63])
    center_slider = Slider(
        ax=ax_center,
        label='Center \nvoxel',
        valmin=0,
        valmax=classification_arr.shape[axis]-1,
        valinit=init_center,
        orientation="vertical"
    )
    center_slider.label.set_size(8)
    ax_depth = fig.add_axes([0.05, 0.1, 0.0225, 0.63])
    depth_slider = Slider(
        ax=ax_depth,
        label="Depth of \nprojection \nin #voxels",
        valmin=1,
        valmax=100,
        valinit=init_depth,
        orientation="vertical"
    )
    depth_slider.label.set_size(8)

    # -- define update function

    def update(val):
        nhit_img, occl_img, lim_ocll, lim_nhit = generate_image(center_slider.val, depth_slider.val, axis)
        im1 = ax.imshow(nhit_img, cmap=greens_cmap, clim=[0.1, lim_nhit], interpolation='none',
                    alpha=1, extent=extent)
        im2 = ax.imshow(occl_img * 100, cmap=reds_cmap, clim=[1, lim_ocll], interpolation='none',
                        alpha=1, extent=extent)

    # --- show

    # plt.show()

    out_file = "test_out/TEMP_slice_fig.png"
    plt.savefig(out_file, dpi=300, format="png", bbox_inches="tight")

o3d_mesh_to_pyvista 

o3d_mesh_to_pyvista(o3d_mesh)

Convert an Open3D triangle mesh to a PyVista PolyData mesh.

Parameters:

  • o3d_mesh (TriangleMesh) –

    Open3D mesh to convert.

Returns:

  • PolyData –

    PyVista mesh with vertices, faces, and optionally vertex colors.

Source code in occpy/visualization.py 
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
def o3d_mesh_to_pyvista(o3d_mesh):
    """
    Convert an Open3D triangle mesh to a PyVista PolyData mesh.

    Parameters
    ----------
    o3d_mesh : open3d.geometry.TriangleMesh
        Open3D mesh to convert.

    Returns
    -------
    pyvista.PolyData
        PyVista mesh with vertices, faces, and optionally vertex colors.
    """

    vertices = np.asarray(o3d_mesh.vertices)
    triangles = np.asarray(o3d_mesh.triangles)
    faces = np.hstack([np.full((triangles.shape[0], 1), 3), triangles]).astype(np.int64).ravel()
    pv_mesh = pv.PolyData(vertices, faces)

    if o3d_mesh.has_vertex_colors():
        colors = np.asarray(o3d_mesh.vertex_colors)
        pv_mesh.point_data["Colors"] = (colors * 255).astype(np.uint8)

    return pv_mesh

plot_riegl_grid 

plot_riegl_grid(data, max_scanline, max_scanline_idx, image2=None, out_path=None)

Plot a scanline-by-index occupancy grid from RIEGL data.

Builds a boolean image with shape (max_scanline_idx+1, max_scanline+1) marking where (scanline, scanline_idx) pairs exist in data. Optionally overlays a second boolean image and saves the figure to out_path.

Parameters:

  • data (DataFrame) –

    DataFrame containing columns 'scanline' and 'scanline_idx'.

  • max_scanline (int) –

    Maximum scanline index on the horizontal axis.

  • max_scanline_idx (int) –

    Maximum scanline_idx on the vertical axis.

  • image2 (array-like of bool, default: None ) –

    Secondary image to overlay (same shape as the grid).

  • out_path (str, default: None ) –

    Path to save the resulting figure.

Source code in occpy/visualization.py 
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
def plot_riegl_grid(data : pd.DataFrame, max_scanline, max_scanline_idx, image2=None, out_path=None):
    """
    Plot a scanline-by-index occupancy grid from RIEGL data.

    Builds a boolean image with shape (max_scanline_idx+1, max_scanline+1) marking
    where (scanline, scanline_idx) pairs exist in `data`. Optionally overlays a
    second boolean image and saves the figure to `out_path`.

    Parameters
    ----------
    data : pandas.DataFrame
        DataFrame containing columns 'scanline' and 'scanline_idx'.
    max_scanline : int
        Maximum scanline index on the horizontal axis.
    max_scanline_idx : int
        Maximum scanline_idx on the vertical axis.
    image2 : array-like of bool, optional
        Secondary image to overlay (same shape as the grid).
    out_path : str, optional
        Path to save the resulting figure.

    """
    scanline_np = data[["scanline"]].to_numpy()
    scanline_idx_np = data[["scanline_idx"]].to_numpy()
    # scanline_idx_np = np.where(scanline_idx_np > max_scanline_idx, max_scanline_idx, scanline_idx_np)
    extent = [0, max_scanline+1, 0, max_scanline_idx+1]
    img = np.zeros(shape=(max_scanline_idx+1, max_scanline+1), dtype=bool)
    img[scanline_idx_np, scanline_np] = True
    figsize=(12,5)
    cmap = matplotlib.colors.ListedColormap(['white', 'red'])
    fig, ax = plt.subplots(ncols=1, nrows=1, squeeze=False, 
                           sharex=False, sharey=False, figsize=figsize)

    with plt.style.context('seaborn-v0_8-notebook'):
        ax[0,0].imshow(img, interpolation='nearest', extent=extent, 
                    clim=[0,1], cmap=plt.get_cmap(cmap, 2), vmin=0, vmax=1, alpha=1)
        ax[0,0].set(adjustable='box', aspect='equal')
        ax[0,0].set(xlabel="Scanline", ylabel="Scanline index")
        ax[0,0].set_facecolor('white')
        cmap_blue = matplotlib.colors.ListedColormap(['white', 'blue'])
        if image2 is not None:
            ax[0,0].imshow(image2, interpolation='nearest', extent=extent, clim=[0,1], cmap=plt.get_cmap(cmap_blue,2), vmin=0,vmax=1, alpha=0.5)
    fig.tight_layout()
    plt.show()
    if out_path is not None:
        fig.savefig(out_path)

vis_pv_interactive 

vis_pv_interactive(occmap_file, min_bound_voxel, max_bound_voxel, config_file, pointcloud_file=None, opacity_occluded=0.2, opacity_hit=0.1, opacity_unobserved=0.2, point_size=2, point_color=(0, 0, 0), return_plotter=True)

Visualize an occupancy map with PyVista in interactive mode.

Loads a voxel occupancy grid (.npy) and optionally point cloud (.las), crops a region based on max_bound and min_bound, and displays occluded (red), hit (green), and unobserved (blue) voxels as semi-transparent meshes, with optionally the point cloud overlaid.

Ensure min_bound and max_bound are within the occupancy grid dimensions. Large regions may be quite slow to render.

Parameters:

  • occmap_file (str) –

    Path to the .npy file containing the occupancy map (3D array).

  • min_bound_voxel (array-like of int, shape (3,)) –

    Minimum XYZ voxel coordinates to visualize

  • max_bound_voxel (array-like of int, shape (3,)) –

    Maximum XYZ voxel coordinates to visualize

  • config_file (str) –

    Path to config file containing occpy run parameters.

  • pointcloud_file (str, default: None ) –

    If provided, visualize the point cloud from this .las file.

  • opacity_occluded (float, default: 0.2 ) –

    Opacity for occluded voxels (red).

  • opacity_hit (float, default: 0.1 ) –

    Opacity for hit voxels (green).

  • opacity_unobserved (float, default: 0.2 ) –

    Opacity for unobserved voxels (blue).

  • point_size (float, default: 2 ) –

    Point size for rendering the point cloud.

  • point_color (tuple of float, default: (0,0,0) ) –

    RGB color for the point cloud points.

  • return_plotter (bool, default: False ) –

    If True, return the PyVista plotter object for further manipulation instead of showing the plot

Source code in occpy/visualization.py 
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
def vis_pv_interactive(occmap_file, 
                       min_bound_voxel, 
                       max_bound_voxel, 
                       config_file,
                       pointcloud_file=None,
                       opacity_occluded=0.2,
                       opacity_hit=0.1,
                       opacity_unobserved=0.2,
                       point_size=2,
                       point_color=(0,0,0),
                       return_plotter=True):
    """
    Visualize an occupancy map with PyVista in interactive mode.

    Loads a voxel occupancy grid (.npy) and optionally point cloud (.las), crops a region based on max_bound and min_bound,
    and displays occluded (red), hit (green), and unobserved (blue) voxels as
    semi-transparent meshes, with optionally the point cloud overlaid.

    Ensure min_bound and max_bound are within the occupancy grid dimensions. Large regions may be quite slow to render.

    Parameters
    ----------
    occmap_file : str
        Path to the .npy file containing the occupancy map (3D array).
    min_bound_voxel : array-like of int, shape (3,)
        Minimum XYZ voxel coordinates to visualize
    max_bound_voxel : array-like of int, shape (3,)
        Maximum XYZ voxel coordinates to visualize
    config_file : str
        Path to config file containing occpy run parameters.
    pointcloud_file : str, default None
        If provided, visualize the point cloud from this .las file.
    opacity_occluded : float, default 0.2
        Opacity for occluded voxels (red).
    opacity_hit : float, default 0.1
        Opacity for hit voxels (green).
    opacity_unobserved : float, default 0.2
        Opacity for unobserved voxels (blue).
    point_size : float, default 2
        Point size for rendering the point cloud.
    point_color : tuple of float, default (0,0,0)
        RGB color for the point cloud points.
    return_plotter : bool, default False
        If True, return the PyVista plotter object for further manipulation instead of showing the plot
    """

    occmap = np.load(occmap_file)
    print("occmap shape:", occmap.shape)

    # read json config file
    with open(config_file) as file:
        settings = json.load(file)

    vox_dim = settings["vox_dim"]
    plot_dim = settings["plot_dim"]
    # get min and max bounds
    min_bound = np.array(plot_dim[:3])
    max_bound = np.array(plot_dim[3:6])

    class_colors = {
        1: np.array([0, 255, 0], dtype=np.uint8),   # hit / green
        3: np.array([255, 0, 0], dtype=np.uint8),   # occluded / red
        4: np.array([0, 0, 255], dtype=np.uint8),   # unobserved / blue
    }
    class_alpha = {
        1: int(np.clip(round(opacity_hit * 255), 0, 255)),
        3: int(np.clip(round(opacity_occluded * 255), 0, 255)),
        4: int(np.clip(round(opacity_unobserved * 255), 0, 255)),
    }

    nx, ny, nz = occmap.shape

    # build full-scale meshes once and keep voxel indices for updates
    def build_mesh_for_class(target_class):
        verts, faces, cell_colors = [], [], []
        voxel_coords = []
        for x in range(nx):
            for y in range(ny):
                for z in range(nz):
                    if occmap[x, y, z] != target_class:
                        continue
                    voxel_coords.append((x, y, z))
                    base = np.array([x, y, z], dtype=float) * vox_dim + min_bound
                    cube_verts = base + vox_dim * np.array([
                        [0,0,0],[1,0,0],[1,1,0],[0,1,0],
                        [0,0,1],[1,0,1],[1,1,1],[0,1,1]
                    ])
                    idx0 = len(verts)
                    verts.extend(cube_verts)
                    cube_faces = [
                        [4, idx0, idx0+1, idx0+2, idx0+3],
                        [4, idx0+4, idx0+5, idx0+6, idx0+7],
                        [4, idx0, idx0+1, idx0+5, idx0+4],
                        [4, idx0+1, idx0+2, idx0+6, idx0+5],
                        [4, idx0+2, idx0+3, idx0+7, idx0+6],
                        [4, idx0+3, idx0, idx0+4, idx0+7],
                    ]
                    faces.extend(cube_faces)
                    cell_colors.extend([class_colors[target_class]]*6)
        verts = np.array(verts)
        faces = np.hstack(faces)
        cell_colors = np.array(cell_colors, dtype=np.uint8)
        # init alpha to 0, window updates will toggle visibility.
        rgba = np.hstack([cell_colors, np.zeros((cell_colors.shape[0],1), dtype=np.uint8)])
        mesh = pv.PolyData(verts, faces)
        mesh.cell_data["rgba"] = rgba
        return mesh, np.array(voxel_coords, dtype=np.int32)

    # build full meshes (slow)
    unique_values = np.unique(occmap)
    mesh_entries = []

    def _add_mesh_entry(mesh, voxels, alpha_on):
        if mesh is None:
            return
        n_cubes = voxels.shape[0]
        axis_bins = [[], [], []]
        for axis, size in enumerate((nx, ny, nz)):
            axis_vals = voxels[:, axis]
            axis_bins[axis] = [np.where(axis_vals == idx)[0] for idx in range(size)]
        mesh_entries.append({
            "mesh": mesh,
            "voxels": voxels,
            "n_cubes": n_cubes,
            "axis_bins": axis_bins,
            "alpha_on": alpha_on,
        })

    print("Constructing meshes, can be slow for large grids")

    if 1 in unique_values:
        mesh_hit, vox_hit = build_mesh_for_class(1)
        _add_mesh_entry(mesh_hit, vox_hit, class_alpha[1])
    else:
        mesh_hit = None
    if 3 in unique_values:
        mesh_occl, vox_occl = build_mesh_for_class(3)
        _add_mesh_entry(mesh_occl, vox_occl, class_alpha[3])
    else:
        mesh_occl = None
    if 4 in unique_values:  
        mesh_unobs, vox_unobs = build_mesh_for_class(4)
        _add_mesh_entry(mesh_unobs, vox_unobs, class_alpha[4])
    else:
        mesh_unobs = None

    print("Mesh construction done")

    # setup plotter
    plotter = pv.Plotter()
    if mesh_hit is not None:
        actor_hit = plotter.add_mesh(mesh_hit, scalars="rgba", rgb=True, lighting=False)
    if mesh_occl is not None:
        actor_occl = plotter.add_mesh(mesh_occl, scalars="rgba", rgb=True, lighting=False)
    if mesh_unobs is not None:
        actor_unobs = plotter.add_mesh(mesh_unobs, scalars="rgba", rgb=True, lighting=False)

    # add point cloud
    point_cloud = None
    point_actor = None
    point_rgba = None
    point_alpha_on = 255
    all_points = None
    points_vox_float = None
    point_axis_bins = None
    point_visible = None

    if pointcloud_file is not None:
        las = laspy.read(pointcloud_file)
        all_points = np.vstack((las.x, las.y, las.z)).transpose()
        points_vox_float = (all_points - min_bound) / vox_dim

        # build bins for fast updates
        points_vox_int = np.floor(points_vox_float).astype(np.int32)
        point_axis_bins = [[], [], []]
        for axis, size in enumerate((nx, ny, nz)):
            axis_vals = points_vox_int[:, axis]
            point_axis_bins[axis] = [np.where(axis_vals == idx)[0] for idx in range(size)]

        point_visible = np.zeros(all_points.shape[0], dtype=bool)

        min_pc_crop = min_bound + np.array([int(min_bound_voxel[0]), int(min_bound_voxel[1]), int(min_bound_voxel[2])]) * vox_dim
        max_pc_crop = min_bound + np.array([int(max_bound_voxel[0]), int(max_bound_voxel[1]), int(max_bound_voxel[2])]) * vox_dim
        init_mask = np.all((all_points >= min_pc_crop) & (all_points <= max_pc_crop), axis=1)
        point_visible[:] = init_mask

        # same logic as mesh: pre allocate points and set visibility with alpha channel
        point_cloud = pv.PolyData(all_points)
        point_rgba = np.zeros((all_points.shape[0], 4), dtype=np.uint8)
        point_rgba[:, :3] = point_color
        point_rgba[:, 3] = np.where(point_visible, point_alpha_on, 0).astype(np.uint8)
        point_cloud["rgba"] = point_rgba
        point_actor = plotter.add_points(
            point_cloud,
            scalars="rgba",
            rgb=True,
            style="points",
            point_size=point_size,
            render_points_as_spheres=True,
        )

    # setup window and updating
    window_x = [int(min_bound_voxel[0]), int(max_bound_voxel[0])]
    window_y = [int(min_bound_voxel[1]), int(max_bound_voxel[1])]
    window_z = [int(min_bound_voxel[2]), int(max_bound_voxel[2])]

    pref_size_x = max(1, window_x[1] - window_x[0])
    pref_size_y = max(1, window_y[1] - window_y[0])
    pref_size_z = max(1, window_z[1] - window_z[0])

    slider_widgets = {}

    bounds_text_actor = None

    def _update_window_text():
        nonlocal bounds_text_actor
        msg = (
            f"Window bounds: X [{window_x[0]}, {window_x[1]}) | "
            f"Y [{window_y[0]}, {window_y[1]}) | "
            f"Z [{window_z[0]}, {window_z[1]})"
        )
        if bounds_text_actor is None:
            bounds_text_actor = plotter.add_text(msg, position="upper_right", font_size=10, name="window_bounds")
        else:
            # update existing actor text
            try:
                bounds_text_actor.SetInput(msg)
            except AttributeError:
                plotter.add_text(msg, position="upper_right", font_size=10, name="window_bounds")

    def _window_mask_for_voxels(voxels):
        return (
            (window_x[0] <= voxels[:, 0]) & (voxels[:, 0] < window_x[1])
            & (window_y[0] <= voxels[:, 1]) & (voxels[:, 1] < window_y[1])
            & (window_z[0] <= voxels[:, 2]) & (voxels[:, 2] < window_z[1])
        )

    def update_window_full():
        nonlocal point_rgba
        for entry in mesh_entries:
            mesh = entry["mesh"]
            voxels = entry["voxels"]
            alpha_on = entry["alpha_on"]
            rgba = mesh.cell_data["rgba"].copy()
            alpha = rgba[:, 3].reshape(-1, 6)
            alpha[:] = 0
            mask = _window_mask_for_voxels(voxels)
            if np.any(mask):
                alpha[mask, :] = alpha_on
            mesh.cell_data["rgba"] = rgba

        if all_points is not None:
            mask = (
                (window_x[0] <= points_vox_float[:, 0]) & (points_vox_float[:, 0] < window_x[1])
                & (window_y[0] <= points_vox_float[:, 1]) & (points_vox_float[:, 1] < window_y[1])
                & (window_z[0] <= points_vox_float[:, 2]) & (points_vox_float[:, 2] < window_z[1])
            )
            point_visible[:] = mask
            point_rgba[:, 3] = np.where(point_visible, point_alpha_on, 0).astype(np.uint8)
            point_cloud["rgba"] = point_rgba

        _update_window_text()

        plotter.render()

    def update_window_incremental(axis, leaving_idx=None, entering_idx=None):
        nonlocal point_rgba
        for entry in mesh_entries:
            mesh = entry["mesh"]
            voxels = entry["voxels"]
            bins = entry["axis_bins"][axis]
            alpha_on = entry["alpha_on"]
            rgba = mesh.cell_data["rgba"].copy()
            alpha = rgba[:, 3].reshape(-1, 6)

            leaving_cubes = bins[leaving_idx] if leaving_idx is not None and 0 <= leaving_idx < len(bins) else np.empty(0, dtype=int)
            if leaving_cubes.size:
                alpha[leaving_cubes, :] = 0

            entering_cubes = bins[entering_idx] if entering_idx is not None and 0 <= entering_idx < len(bins) else np.empty(0, dtype=int)
            if entering_cubes.size:
                entering_voxels = voxels[entering_cubes]
                visible_mask = _window_mask_for_voxels(entering_voxels)
                if np.any(visible_mask):
                    alpha[entering_cubes[visible_mask], :] = alpha_on

            mesh.cell_data["rgba"] = rgba

        if all_points is not None:
            if leaving_idx is not None and 0 <= leaving_idx < len(point_axis_bins[axis]):
                leaving_points = point_axis_bins[axis][leaving_idx]
                if leaving_points.size:
                    point_visible[leaving_points] = False

            if entering_idx is not None and 0 <= entering_idx < len(point_axis_bins[axis]):
                entering_points = point_axis_bins[axis][entering_idx]
                if entering_points.size:
                    v = points_vox_float[entering_points]
                    keep = (
                        (window_x[0] <= v[:, 0]) & (v[:, 0] < window_x[1])
                        & (window_y[0] <= v[:, 1]) & (v[:, 1] < window_y[1])
                        & (window_z[0] <= v[:, 2]) & (v[:, 2] < window_z[1])
                    )
                    point_visible[entering_points] = keep

            point_rgba[:, 3] = np.where(point_visible, point_alpha_on, 0).astype(np.uint8)
            point_cloud["rgba"] = point_rgba

        _update_window_text()

        plotter.render()

    def _axis_window_and_dim(axis):
        if axis == 0:
            return window_x, nx
        if axis == 1:
            return window_y, ny
        return window_z, nz

    def _set_slider_representation(axis, value):
        widget = slider_widgets.get(axis)
        if widget is None:
            return
        rep = widget.GetSliderRepresentation()
        rep.SetValue(int(value))
        rep.SetLabelFormat('%.0f')

    def _set_window_start(axis, start_value):
        win, dim = _axis_window_and_dim(axis)
        if axis == 0:
            pref_size = pref_size_x
        elif axis == 1:
            pref_size = pref_size_y
        else:
            pref_size = pref_size_z

        # Sliders represent absolute indices [0, dim-1].
        # shrink at boundaries
        start = int(round(start_value))
        start = max(0, min(start, dim - 1))

        new_start = start
        new_end = min(new_start + pref_size, dim)

        old_start = win[0]
        old_end = win[1]

        if new_start == old_start and new_end == old_end:
            _set_slider_representation(axis, new_start)
            return

        win[0] = new_start
        win[1] = new_end

        # incremental shift using bins if possible
        if new_start == old_start + 1 and new_end == old_end + 1:
            update_window_incremental(axis=axis, leaving_idx=old_start, entering_idx=old_end)
        elif new_start == old_start + 1 and new_end == old_end:
            update_window_incremental(axis=axis, leaving_idx=old_start, entering_idx=None)
        elif new_start == old_start - 1 and new_end == old_end - 1:
            update_window_incremental(axis=axis, leaving_idx=old_end - 1, entering_idx=old_start - 1)
        elif new_start == old_start - 1 and new_end == old_end:
            update_window_incremental(axis=axis, leaving_idx=None, entering_idx=old_start - 1)
        else:
            update_window_full()

        _set_slider_representation(axis, new_start)

    def move_window(axis, step):
        win, _ = _axis_window_and_dim(axis)
        _set_window_start(axis, win[0] + int(step))

    # on-screen controls for notebook where key events don't work
    win_size_x = window_x[1] - window_x[0]
    win_size_y = window_y[1] - window_y[0]
    win_size_z = window_z[1] - window_z[0]

    plotter.add_text(
        "Window Controls: sliders (x/y/z) | Keys: x/b, y/n, z/m",
        position="upper_left",
        font_size=10,
    )

    def _slider_callback(value, widget, axis):
        discrete_value = int(round(value))
        _set_window_start(axis, discrete_value)
        rep = widget.GetSliderRepresentation()
        rep.SetValue(int(round(discrete_value)))
        rep.SetLabelFormat('%.0f')

    slider_widgets[0] = plotter.add_slider_widget(
        callback=partial(_slider_callback, axis=0),
        rng=[0, max(0, nx - 1)],
        value=window_x[0],
        title="X start",
        pointa=(0.03, 0.04),
        pointb=(0.33, 0.04),
        pass_widget=True,
        fmt="%.0f",
    )
    slider_widgets[1] = plotter.add_slider_widget(
        callback=partial(_slider_callback, axis=1),
        rng=[0, max(0, ny - 1)],
        value=window_y[0],
        title="Y start",
        pointa=(0.35, 0.04),
        pointb=(0.65, 0.04),
        pass_widget=True,
        fmt="%.0f",
    )
    slider_widgets[2] = plotter.add_slider_widget(
        callback=partial(_slider_callback, axis=2),
        rng=[0, max(0, nz - 1)],
        value=window_z[0],
        title="Z start",
        pointa=(0.67, 0.04),
        pointb=(0.97, 0.04),
        pass_widget=True,
        fmt="%.0f",
    )

    # key callbacks
    plotter.add_key_event("x", lambda: move_window(axis=0, step=1))
    plotter.add_key_event("b", lambda: move_window(axis=0, step=-1))
    plotter.add_key_event("y", lambda: move_window(axis=1, step=1))
    plotter.add_key_event("n", lambda: move_window(axis=1, step=-1))
    plotter.add_key_event("z", lambda: move_window(axis=2, step=1))
    plotter.add_key_event("m", lambda: move_window(axis=2, step=-1))

    # init window
    update_window_full()

    if return_plotter:
        return plotter
    else:
        plotter.show()

vis_pv_rotating 

vis_pv_rotating(occmap_file, min_bound_voxel, max_bound_voxel, config_file, opath='occpy_pv_rotating.mp4', pointcloud_file=None, opacity_occluded=0.2, opacity_hit=0.1, opacity_unobserved=0.2, point_size=2, point_color=(0, 0, 0), framerate=30, n_frames=180, distance_factor=2, camera_elevation=30)

Create a rotating visualization of an occupancy map with PyVista.

Loads a voxel occupancy grid (.npy) and optionally point cloud (.las), crops a region based on max_bound and min_bound, and displays occluded (red), hit (green), and unobserved (blue) voxels as semi-transparent meshes, with optionally the point cloud overlaid.

Ensure min_bound and max_bound are within the occupancy grid dimensions. Large regions may be quite slow to render.

Parameters:

  • occmap_file (str) –

    Path to the .npy file containing the occupancy map (3D array).

  • min_bound_voxel (array-like of float, shape (3,)) –

    Minimum XYZ voxel coordinates to visualize

  • max_bound_voxel (array-like of float, shape (3,)) –

    Maximum XYZ voxel coordinates to visualize

  • config_file (str) –

    Path to config file containing occpy run parameters.

  • opath –

    Output path for the rotating video.

  • pointcloud_file (str, default: None ) –

    If provided, visualize the point cloud from this .las file.

  • opacity_occluded (float, default: 0.2 ) –

    Opacity for occluded voxels (red).

  • opacity_hit (float, default: 0.1 ) –

    Opacity for hit voxels (green).

  • opacity_unobserved (float, default: 0.2 ) –

    Opacity for unobserved voxels (blue).

  • point_size (float, default: 2 ) –

    Point size for rendering the point cloud.

  • point_color (tuple of float, default: (0,0,0) ) –

    RGB color for the point cloud points.

  • framerate (int, default: 30 ) –

    Frames per second for the output video.

  • n_frames (int, default: 180 ) –

    Number of frames in the rotation (e.g. 180 for a full 360 degree rotation at 2 degrees per frame).

  • distance_factor –

    Controls distance to scene for camera orbit. Higher values will show more of the scene but may reduce visibility of details.

  • camera_elevation –

    Elevation angle of the camera in degrees. Higher values will show more of the top-down view, lower values will be more level with the scene.

Source code in occpy/visualization.py 
 889
 890
 891
 892
 893
 894
 895
 896
 897
 898
 899
 900
 901
 902
 903
 904
 905
 906
 907
 908
 909
 910
 911
 912
 913
 914
 915
 916
 917
 918
 919
 920
 921
 922
 923
 924
 925
 926
 927
 928
 929
 930
 931
 932
 933
 934
 935
 936
 937
 938
 939
 940
 941
 942
 943
 944
 945
 946
 947
 948
 949
 950
 951
 952
 953
 954
 955
 956
 957
 958
 959
 960
 961
 962
 963
 964
 965
 966
 967
 968
 969
 970
 971
 972
 973
 974
 975
 976
 977
 978
 979
 980
 981
 982
 983
 984
 985
 986
 987
 988
 989
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
def vis_pv_rotating(occmap_file, 
                min_bound_voxel, 
                max_bound_voxel, 
                config_file,
                opath="occpy_pv_rotating.mp4",
                pointcloud_file=None,
                opacity_occluded=0.2,
                opacity_hit=0.1,
                opacity_unobserved=0.2,
                point_size=2,
                point_color=(0,0,0),
                framerate=30,
                n_frames=180,
                distance_factor=2,
                camera_elevation=30):
    """
    Create a rotating visualization of an occupancy map with PyVista.

    Loads a voxel occupancy grid (.npy) and optionally point cloud (.las), crops a region based on max_bound and min_bound,
    and displays occluded (red), hit (green), and unobserved (blue) voxels as
    semi-transparent meshes, with optionally the point cloud overlaid.

    Ensure min_bound and max_bound are within the occupancy grid dimensions. Large regions may be quite slow to render.

    Parameters
    ----------
    occmap_file : str
        Path to the .npy file containing the occupancy map (3D array).
    min_bound_voxel : array-like of float, shape (3,)
        Minimum XYZ voxel coordinates to visualize
    max_bound_voxel : array-like of float, shape (3,)
        Maximum XYZ voxel coordinates to visualize
    config_file : str
        Path to config file containing occpy run parameters.
    opath: str, default "occpy_pv_rotating.mp4"
        Output path for the rotating video.
    pointcloud_file : str, default None
        If provided, visualize the point cloud from this .las file.
    opacity_occluded : float, default 0.2
        Opacity for occluded voxels (red).
    opacity_hit : float, default 0.1
        Opacity for hit voxels (green).
    opacity_unobserved : float, default 0.2
        Opacity for unobserved voxels (blue).
    point_size : float, default 2
        Point size for rendering the point cloud.
    point_color : tuple of float, default (0,0,0)
        RGB color for the point cloud points.
    framerate : int, default 30
        Frames per second for the output video.
    n_frames : int, default 180
        Number of frames in the rotation (e.g. 180 for a full 360 degree rotation at 2 degrees per frame).
    distance_factor: float, default 1.5
        Controls distance to scene for camera orbit. Higher values will show more of the scene but may reduce visibility of details.
    camera_elevation: int, default 30
        Elevation angle of the camera in degrees. Higher values will show more of the top-down view, lower values will be more level with the scene.
    """

    if not os.path.exists(occmap_file):
        raise FileNotFoundError(f"Occupancy map file not found: {occmap_file}")
    if not os.path.exists(config_file):
        raise FileNotFoundError(f"Config file not found: {config_file}")
    if pointcloud_file is not None and not os.path.exists(pointcloud_file):
        raise FileNotFoundError(f"Point cloud file not found: {pointcloud_file}")

    # read json config file
    with open(config_file) as file:
        settings = json.load(file)

    if "vox_dim" not in settings or "plot_dim" not in settings:
        raise ValueError(f"Config file must contain 'vox_dim' and 'plot_dim' keys. Found keys: {list(settings.keys())}")

    # check if min_bound_voxel and max_bound_voxel are valid
    if not all(isinstance(x, int) for x in min_bound_voxel) or not all(isinstance(x, int) for x in max_bound_voxel):
        raise ValueError(f"min_bound_voxel and max_bound_voxel must be array-like of int. Got min_bound_voxel={min_bound_voxel}, max_bound_voxel={max_bound_voxel}")
    min_bound_voxel = np.array(min_bound_voxel)
    max_bound_voxel = np.array(max_bound_voxel)

    vox_dim = settings["vox_dim"]
    plot_dim = settings["plot_dim"]
    # get min and max bounds
    min_bound = np.array(plot_dim[:3])
    max_bound = np.array(plot_dim[3:6])

    occmap = np.load(occmap_file)
    dims = occmap.shape

    # check if min_bound and max_bound are inside dims
    for i in range(3):
        if min_bound_voxel[i] < 0 or max_bound_voxel[i] > dims[i]:
            raise ValueError(f"min_bound and max_bound must be within occupancy map dimensions {dims}, got min_bound_voxel={min_bound_voxel}, max_bound_voxel={max_bound_voxel}")

    # collect bounding boxes for each voxel type
    bboxs_occl = []
    bbox_unobserved = []
    bbox_hit = []

    for x in range(min_bound_voxel[0], max_bound_voxel[0]):
        for y in range(min_bound_voxel[1], max_bound_voxel[1]):
            for z in range(min_bound_voxel[2], max_bound_voxel[2]):
                min_bound_cube = min_bound + np.array([x, y, z]) * vox_dim
                max_bound_cube = min_bound_cube + vox_dim

                if occmap[x, y, z] == 3:  # Occluded
                    bboxs_occl.append(o3d.geometry.AxisAlignedBoundingBox(min_bound_cube, max_bound_cube))
                elif occmap[x, y, z] == 4:  # Unobserved
                    bbox_unobserved.append(o3d.geometry.AxisAlignedBoundingBox(min_bound_cube, max_bound_cube))
                elif occmap[x, y, z] == 1:  # Hit
                    bbox_hit.append(o3d.geometry.AxisAlignedBoundingBox(min_bound_cube, max_bound_cube))

    # create meshes
    plotter = pv.Plotter(notebook=False, off_screen=True, window_size=[2048, 2048])

    if len(bboxs_occl) > 0:
        mesh_occl = batch_aabbs_to_mesh(bboxs_occl)
        pv_mesh_occl = o3d_mesh_to_pyvista(mesh_occl)
        plotter.add_mesh(pv_mesh_occl, opacity=opacity_occluded, color='red')

    if len(bbox_hit) > 0:
        mesh_hit = batch_aabbs_to_mesh(bbox_hit)
        pv_mesh_hit = o3d_mesh_to_pyvista(mesh_hit)
        plotter.add_mesh(pv_mesh_hit, opacity=opacity_hit, color='green')

    if len(bbox_unobserved) > 0:
        mesh_unobserved = batch_aabbs_to_mesh(bbox_unobserved)
        pv_mesh_unobserved = o3d_mesh_to_pyvista(mesh_unobserved)
        plotter.add_mesh(pv_mesh_unobserved, opacity=opacity_unobserved, color='blue')

    # read and add point cloud if given
    if pointcloud_file is not None:
        las = laspy.read(pointcloud_file)
        points = np.vstack((las.x, las.y, las.z)).transpose()
        min_pc_crop = min_bound + min_bound_voxel * vox_dim
        max_pc_crop = min_bound + max_bound_voxel * vox_dim
        print(f"Crop bounds: {min_pc_crop} to {max_pc_crop}")

        mask = np.all((points >= min_pc_crop) & (points <= max_pc_crop), axis=1)
        points_in_crop = points[mask]
        if len(points_in_crop) == 0:
            print("Warning: No points in the point cloud are within the crop bounds.")
        else:
            point_cloud = pv.PolyData(points_in_crop)
            point_cloud["point_color"] = np.repeat(point_color, len(points_in_crop), axis=0).reshape(-1, 3)
            plotter.add_points(point_cloud, scalars='point_color', style="points", point_size=point_size, render_points_as_spheres=True)
            plotter.remove_scalar_bar()

    # calculate center and radius for camera orbit
    min_bound_voxel = np.array(min_bound_voxel)
    max_bound_voxel = np.array(max_bound_voxel)
    center_voxel = (min_bound_voxel + max_bound_voxel) / 2
    bounds_size_voxel = max_bound_voxel - min_bound_voxel
    center_world = min_bound + center_voxel * vox_dim
    radius = np.linalg.norm(bounds_size_voxel * vox_dim) * distance_factor
    plotter.open_movie(opath, framerate=framerate)

    for i in range(n_frames):
        azimuth = i * (360 / n_frames)

        plotter.camera.position = (
            center_world[0] + radius * np.cos(np.radians(azimuth)) * np.cos(np.radians(camera_elevation)),
            center_world[1] + radius * np.sin(np.radians(azimuth)) * np.cos(np.radians(camera_elevation)),
            center_world[2] + radius * np.sin(np.radians(camera_elevation))
        )

        plotter.camera.focal_point = center_world
        plotter.camera.up = (0, 0, 1)

        plotter.write_frame()

    plotter.close()

vis_pv_static_bounds 

vis_pv_static_bounds(occmap_file, min_bound_voxel, max_bound_voxel, config_file, pointcloud_file=None, opacity_occluded=0.2, opacity_hit=0.1, opacity_unobserved=0.2, point_size=2, point_color=(0, 0, 0), return_plotter=True)

Visualize an occupancy map with PyVista.

Loads a voxel occupancy grid (.npy) and optionally point cloud (.las), crops a region based on max_bound and min_bound, and displays occluded (red), hit (green), and unobserved (blue) voxels as semi-transparent meshes, with optionally the point cloud overlaid.

Ensure min_bound and max_bound are within the occupancy grid dimensions. Large regions may be quite slow to render.

Parameters:

  • occmap_file (str) –

    Path to the .npy file containing the occupancy map (3D array).

  • min_bound_voxel (array-like of int, shape (3,)) –

    Minimum XYZ voxel coordinates to visualize

  • max_bound_voxel (array-like of int, shape (3,)) –

    Maximum XYZ voxel coordinates to visualize

  • config_file (str) –

    Path to config file containing occpy run parameters.

  • pointcloud_file (str, default: None ) –

    If provided, visualize the point cloud from this .las file.

  • opacity_occluded (float, default: 0.2 ) –

    Opacity for occluded voxels (red).

  • opacity_hit (float, default: 0.1 ) –

    Opacity for hit voxels (green).

  • opacity_unobserved (float, default: 0.2 ) –

    Opacity for unobserved voxels (blue).

  • point_size (float, default: 2 ) –

    Point size for rendering the point cloud.

  • point_color (tuple of float, default: (0,0,0) ) –

    RGB color for the point cloud points.

  • return_plotter (bool, default: False ) –

    If True, return the PyVista plotter object for further manipulation instead of showing the plot

Source code in occpy/visualization.py 
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
def vis_pv_static_bounds(occmap_file, 
                       min_bound_voxel, 
                       max_bound_voxel, 
                       config_file,
                       pointcloud_file=None,
                       opacity_occluded=0.2,
                       opacity_hit=0.1,
                       opacity_unobserved=0.2,
                       point_size=2,
                       point_color=(0,0,0),
                       return_plotter=True):

    """
    Visualize an occupancy map with PyVista.

    Loads a voxel occupancy grid (.npy) and optionally point cloud (.las), crops a region based on max_bound and min_bound,
    and displays occluded (red), hit (green), and unobserved (blue) voxels as
    semi-transparent meshes, with optionally the point cloud overlaid.

    Ensure min_bound and max_bound are within the occupancy grid dimensions. Large regions may be quite slow to render.

    Parameters
    ----------
    occmap_file : str
        Path to the .npy file containing the occupancy map (3D array).
    min_bound_voxel : array-like of int, shape (3,)
        Minimum XYZ voxel coordinates to visualize
    max_bound_voxel : array-like of int, shape (3,)
        Maximum XYZ voxel coordinates to visualize
    config_file : str
        Path to config file containing occpy run parameters.
    pointcloud_file : str, default None
        If provided, visualize the point cloud from this .las file.
    opacity_occluded : float, default 0.2
        Opacity for occluded voxels (red).
    opacity_hit : float, default 0.1
        Opacity for hit voxels (green).
    opacity_unobserved : float, default 0.2
        Opacity for unobserved voxels (blue).
    point_size : float, default 2
        Point size for rendering the point cloud.
    point_color : tuple of float, default (0,0,0)
        RGB color for the point cloud points.
    return_plotter : bool, default False
        If True, return the PyVista plotter object for further manipulation instead of showing the plot
    """

    # check paths
    if not os.path.exists(occmap_file):
        raise FileNotFoundError(f"Occupancy map file not found: {occmap_file}")
    if not os.path.exists(config_file):
        raise FileNotFoundError(f"Config file not found: {config_file}")
    if pointcloud_file is not None and not os.path.exists(pointcloud_file):
        raise FileNotFoundError(f"Point cloud file not found: {pointcloud_file}")

    # read json config file
    with open(config_file) as file:
        settings = json.load(file)

    # check required keys
    if "vox_dim" not in settings or "plot_dim" not in settings:
        raise ValueError(f"Config file must contain 'vox_dim' and 'plot_dim' keys. Found keys: {list(settings.keys())}")

    # check if min_bound_voxel and max_bound_voxel are valid
    if not all(isinstance(x, int) for x in min_bound_voxel) or not all(isinstance(x, int) for x in max_bound_voxel):
        raise ValueError(f"min_bound_voxel and max_bound_voxel must be array-like of int. Got min_bound_voxel={min_bound_voxel}, max_bound_voxel={max_bound_voxel}")
    min_bound_voxel = np.array(min_bound_voxel)
    max_bound_voxel = np.array(max_bound_voxel)

    vox_dim = settings["vox_dim"]
    plot_dim = settings["plot_dim"]
    # get min and max bounds
    min_bound = np.array(plot_dim[:3])
    max_bound = np.array(plot_dim[3:6])

    occmap = np.load(occmap_file)
    dims = occmap.shape

    print(f"Occlusion map dimensions: {dims}")

    # check if min_bound and max_bound are inside dims
    for i in range(3):
        if min_bound_voxel[i] < 0 or max_bound_voxel[i] > dims[i]:
            raise ValueError(f"min_bound and max_bound must be within occupancy map dimensions {dims}, got min_bound_voxel={min_bound_voxel}, max_bound_voxel={max_bound_voxel}")

    # collect bounding boxes for each voxel type
    bboxs_occl = []
    bbox_unobserved = []
    bbox_hit = []

    for x in range(min_bound_voxel[0], max_bound_voxel[0]):
        for y in range(min_bound_voxel[1], max_bound_voxel[1]):
            for z in range(min_bound_voxel[2], max_bound_voxel[2]):
                min_bound_cube = min_bound + np.array([x, y, z]) * vox_dim
                max_bound_cube = min_bound_cube + vox_dim

                if occmap[x, y, z] == 3:  # Occluded
                    bboxs_occl.append(o3d.geometry.AxisAlignedBoundingBox(min_bound_cube, max_bound_cube))
                elif occmap[x, y, z] == 4:  # Unobserved
                    bbox_unobserved.append(o3d.geometry.AxisAlignedBoundingBox(min_bound_cube, max_bound_cube))
                elif occmap[x, y, z] == 1:  # Hit
                    bbox_hit.append(o3d.geometry.AxisAlignedBoundingBox(min_bound_cube, max_bound_cube))

    # create meshes
    plotter = pv.Plotter()

    if len(bboxs_occl) > 0:
        mesh_occl = batch_aabbs_to_mesh(bboxs_occl)
        pv_mesh_occl = o3d_mesh_to_pyvista(mesh_occl)
        plotter.add_mesh(pv_mesh_occl, opacity=opacity_occluded, color='red')

    if len(bbox_hit) > 0:
        mesh_hit = batch_aabbs_to_mesh(bbox_hit)
        pv_mesh_hit = o3d_mesh_to_pyvista(mesh_hit)
        plotter.add_mesh(pv_mesh_hit, opacity=opacity_hit, color='green')

    if len(bbox_unobserved) > 0:
        mesh_unobserved = batch_aabbs_to_mesh(bbox_unobserved)
        pv_mesh_unobserved = o3d_mesh_to_pyvista(mesh_unobserved)
        plotter.add_mesh(pv_mesh_unobserved, opacity=opacity_unobserved, color='blue')

    # add point cloud (cropped to visible region)
    if pointcloud_file is not None:
        las = laspy.read(pointcloud_file)
        points = np.vstack((las.x, las.y, las.z)).transpose()
        min_pc_crop = min_bound + min_bound_voxel * vox_dim
        max_pc_crop = min_bound + max_bound_voxel * vox_dim
        print(f"Crop bounds: {min_pc_crop} to {max_pc_crop}")

        mask = np.all((points >= min_pc_crop) & (points <= max_pc_crop), axis=1)
        points_in_crop = points[mask]
        if len(points_in_crop) == 0:
            print("Warning: No points in the point cloud are within the crop bounds.")
        else:
            point_cloud = pv.PolyData(points_in_crop)
            point_cloud["point_color"] = np.repeat(point_color, len(points_in_crop), axis=0).reshape(-1, 3)
            plotter.add_points(point_cloud, scalars='point_color', style="points", point_size=point_size, render_points_as_spheres=True)
            plotter.remove_scalar_bar()

    if return_plotter:
        return plotter
    else:
        plotter.show()