TeFlow: Enabling Multi-frame Supervision for Self-Supervised Feed-forward Scene Flow Estimation

ArXi:2602.19053v2 Announce Type: replace Self-supervised feed-forward methods for scene flow estimation offer real-time efficiency, but their supervision from two-frame point correspondences is unreliable and often breaks down under occlusions. Multi-frame supervision has the potential to provide stable guidance by incorporating motion cues from past frames, yet naive extensions of two-frame objectives are ineffective because point correspondences vary abruptly across frames, producing inconsistent signals.