wondeful work, but a dumb question

[lucasjinreal]

TensoRF looks like small and very fast traning, is that possible to make the final RGBD result merged into a single point cloud? If so, how? that would be very helpful if can generates a final point cloud.

[apchenstu]

Yeah, point cloud sometimes is really useful, actually, you could convert the depth maps and the color result into point could in a simple way from here. You could convert them into point cloud by p = ray_o + depth*ray_d.

[UPstartDeveloper]

Hi @jinfagang

I was planning to try this out too - as another option, it looks like it is possible to visualize a point cloud directly from RGBD scans if you use open3d: docs link.

I reckon the only pre-requisite to this approach is we would need the depth map to be a grayscale image, and saved separately from the rgb_map.