Some questions about ba
Excuse me, I noticed that you used triangulated feature points when performing BA optimization in vggsfm. I want to know why the triangulated points are not in the same position as the point cloud obtained by the original vggt. Did you do anything with the depth? I migrated the BA process of vggsfm to vggt, using the camera parameters predicted by vggt as the initial values and the points extracted by sift as feature points.
Hi, I am a little confused about this question. Could you explain it in detail?
Processing non visible frames: [4]
Not enough inliers per frame, skip BA.
Traceback (most recent call last):
File "D:\PYCHARM\PyCharm 2024.1\VGGT\vggt-main\demo_colmap.py", line 297, in
Hello, when I was running the following code, I encountered the aforementioned issue. Could you please advise on how to solve it? Thank you for your response. python demo_colmap.py --scene_dir="D:/3dstruction/SparseRecon/SparseRecon/vggt/examples/room/" --use_ba --max_query_pts=2048 --query_frame_num=5