MESSENGER: Memory-Enhanced Sequential Scene Flow Estimation via Autoregressive Next-Frame Forecasting

Key points

• Early pairwise estimators relying on instantaneous two-frame motion lack long-term temporal correlation and also struggle with poor extrapolation ability in future prediction.1
• Although some recent methods attempt to explore multi-frame scene flow estimation in a sequence-to-sequence manner, they typically suffer from heavy computational overhead with increasing input frames and long-horizon prediction degradation due to ineffective motion propagation.1
• To address these problems, we propose a novel memory-enhanced sequential scene flow pipeline, called MESSENGER.1
• To sufficiently mine long-term temporal dependencies naturally within consecutive sequences, a memory buffer is designed by explicitly storing multiple history flow estimates and latent states.1

Sources (1)

• [1]MESSENGER: Memory-Enhanced Sequential Scene Flow Estimation via Autoregressive Next-Frame Forecasting
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 06:23 PM

  Scene flow can capture low-level 3D motion displacements in dynamic scenarios.

  Early pairwise estimators relying on instantaneous two-frame motion lack long-term temporal correlation and also struggle with poor extrapolation ability in future prediction.

Extractive summary: sentences quoted from the sources.