Unable to use my own point cloud and camera poses
Opened this issue · 0 comments
engs2570 commented
Unable to use my own point cloud and camera pose.
I am trying to splatfacto. However, I do not want to use colmap. I have my own camera parameters and point cloud for each image. These are ground truth parameters. The colmap prediction is obviously not as good as ground truth. Consequently the reconstruction is not satisfactory. This I want to understand how to use my own parameters to train splat-facto.