Open SuzyZhao0417 opened 1 year ago
Hi! Thanks for your kind words!
In the first static scene reconstruction stage, we follow the LLFF code (https://github.com/Fyusion/LLFF) to preprocess the calibrated results generated using COLMAP. This step will generate the poses_bounds.npy. It mainly contains the camera poses and bounds of the scene.
Then, to ensure the same camera poses and bounds for the second few-view dynamic scene reconstruction stage, we store the poses and bounds for the few view camera views when we use the poses_bounds.npy in the first stage. This step will generate the poses.npy and bds.npy.
Hope that clarifies. Thank you!
Hi! Thanks for your great work. I want to test my own data but I don't know how to get bds.npy, could you please tell me what's about?