You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, since our supervision target is the depth map, we directly use the depth map for TSDF fusion to extract the mesh. From my personal experience, depth maps are more prone to noise and holes, and the geometric details are slightly inferior, but the extraction cost is low. On the other hand, extracting the mesh directly in 3D is more robust to noise and potentially offers better geometric details, but the extraction cost might be higher.
Why choose TSDF method to extract mesh? Which method of extracting mesh is better?
The text was updated successfully, but these errors were encountered: