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
Thanks for the impressive and excellent work! After I run the demo, I noticed that the intrinsics of camera sequence from one video is not same using the function (estimate_focal_knowing_depth) from depth, how to get a shared focal of all cameras for all frames.
Hi @tommaoer Thank you for your interest! Intrinsics are modeled implicitly and solved from pointmaps, so they aren't forced to be same across frames. VGGT has implemented a similar function. You can check this issue for more details on recovering shared intrinsics.
Activity
fanegg commentedon Dec 5, 2025
Hi @tommaoer Thank you for your interest! Intrinsics are modeled implicitly and solved from pointmaps, so they aren't forced to be same across frames. VGGT has implemented a similar function. You can check this issue for more details on recovering shared intrinsics.