Ray Casting in a Voxel Block Grid¶

Note

This is NOT ray casting for triangle meshes. Please refer to open3d.t.geometry.RaycastingScene for that use case.

Ray casting can be performed in a voxel block grid to generate depth and color images at specific view points without extracting the entire surface. It is useful for frame-to-model tracking, and for differentiable volume rendering.

We provide optimized conventional rendering, and basic support for customized rendering that may be used in differentiable rendering. An example can be found at examples/python/t_reconstruction_system/ray_casting.py.

Conventional rendering¶

From a reconstructed voxel block grid from TSDF Integration, we can efficiently render the scene given the input depth as a rough range estimate.

 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 # examples/python/t_reconstruction_system/ray_casting.py result = vbg.ray_cast(block_coords=frustum_block_coords, intrinsic=intrinsic, extrinsic=extrinsic, width=depth.columns, height=depth.rows, render_attributes=[ 'depth', 'normal', 'color', 'index', 'interp_ratio' ], depth_scale=config.depth_scale, depth_min=config.depth_min, depth_max=config.depth_max, weight_threshold=1, range_map_down_factor=8) fig, axs = plt.subplots(2, 2) 

The results could be directly obtained and visualized by

  90 91 92 93 94 95 96 97 98 99 100 101 102 # examples/python/t_reconstruction_system/ray_casting.py # Colorized depth colorized_depth = o3d.t.geometry.Image(result['depth']).colorize_depth( axs[0, 0].imshow(colorized_depth.as_tensor().cpu().numpy()) axs[0, 0].set_title('depth') axs[0, 1].imshow(result['normal'].cpu().numpy()) axs[0, 1].set_title('normal') axs[1, 0].imshow(result['color'].cpu().numpy()) 

Customized rendering¶

In customized rendering, we manually perform trilinear-interpolation by accessing properties at 8 nearest neighbor voxels with respect to the found surface point per pixel:

  97 98 99 100 101 102 103 104 105 # examples/python/t_reconstruction_system/ray_casting.py # Render color via indexing vbg_color = vbg.attribute('color').reshape((-1, 3)) nb_indices = result['index'].reshape((-1)) nb_interp_ratio = result['interp_ratio'].reshape((-1, 1)) nb_colors = vbg_color[nb_indices] * nb_interp_ratio sum_colors = nb_colors.reshape((depth.rows, depth.columns, 8, 3)).sum( axs[1, 1].imshow(sum_colors.cpu().numpy()) 

Since the output is rendered via indices, the rendering process could be rewritten in differentiable engines like PyTorch seamlessly via PyTorch I/O with DLPack memory map.