Jul 2026
Event3R: Asynchronous-to-Global 3D Reconstruction from Event Camera via Spatial-Temporal Feature Aggregation
This work introduces Event3R, a feed-forward framework that directly maps asynchronous event streams to globally consistent 3D point clouds, and proposes a Masked Bin Modeling strategy for self-supervised pre-training, enabling robust temporal representation learning with minimal labeled data.
Jian Huang, Haotian Shen, Xinhao Lou et al.
· arXiv.org · 0 citations