Coordinate frame and other questions

[wamiq-reyaz]

Hiya,

You guys have released code that deals with the DTU dataset. Could you please clarify the global coordinate frame and the camera coordinates that the code expects? This would be a great boon in training on our own datasets - especially the synthetic nerf dataset.

So the questions would be in the world coordinates is z-up? In the camera coordinates, what are the coordinate ranges after perspective transform (-1 to 1, or 0 to 1). In the camera, is +z the forward direction or is it -z?

Other question are:

1. Why is there no object_bounding_radius? it is set for nuemesh, but the get_neumesh_model expects it under data?
2. Why do you set perturb to False?
3. What is white_background?

Hope you can answer. Cheers, Wamiq

[chobao]

Hi, wamiq-reyaz! The code dealing with DTU dataset is borrowed from neurecon and the processed DTU data is borrowed from Neus. You can refer to them for more details.

• As far as I know, the camera coordinate is an opencv camera model (i.e. right: x-axis, down: y-axis).

• The backbone of our method is Neus, which expects that the object is within a unit sphere (i.e. obj_bounding_radius). Therefore the coordinate usually ranges from -1 to 1.

• The white background is controlled by white_bkgd in volume_render.

[chobao]

Close the inactive issues. Feel free to reopen it if the problem still needs to be solved.