search

Totoro97 / f2-nerf

Fast neural radiance field training with free camera trajectories
https://totoro97.github.io/projects/f2-nerf/
Apache License 2.0
933 stars 69 forks source link

camera pose normalization #39

Open ETbaty opened 1 year ago

ETbaty commented 1 year ago

Hi, thanks for publishing code for this great work! I see that you normalize the camera poses in the init of dataset. If I want to retrieve point cloud from depth maps from trained scenes is it possible to get it in the original pose before normalization ? so for example the point cloud will be to scale to the original calibration.

Totoro97 commented 1 year ago

Hi, thanks for your interest in this project. The normalization here is simple: Take the average position of the cameras positions as p and the max distance from cameras to p as r, the normalization operation for point x is x = (x - p) / r.