A feed-forward network predicts two time-aware pointmaps per frame pair to simultaneously reconstruct dynamic scenes and track 3D points in a consistent world frame.
TAP-vid: A benchmark for track- ing any point in a video
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
St4RTrack: Simultaneous 4D Reconstruction and Tracking in the World
A feed-forward network predicts two time-aware pointmaps per frame pair to simultaneously reconstruct dynamic scenes and track 3D points in a consistent world frame.