Reproduce “Fast-Slow Evolutionary Occupancy Prediction via Controlled Dynamics” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Runhe Yang, Zhiyuan Zhou, Yuxiang Yan, Xuetong Yang, Yikai Pan, Jiacheng Tang, Zhuolin He, Jianhua Han and 2 more

EvoOcc updates a fast 3D semantic occupancy prediction using delayed outputs from a stronger predictor. It encodes differences between fast and slow predictions, then warps state for vehicle motion and uses changes in fast features to evolve that state. On streaming Occ3D-nuScenes, one pair reaches 40.3 mIoU, exceeding the fast baseline; results depend on predictor quality and exclude communication delay.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.