Reproduce “Surround-View Vision-based 3D Detection for Autonomous Driving: A Survey” 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
Apoorv Singh, Varun Bankiti
The authors survey camera-based 3D detection methods that map calibrated surround-view images into a shared bird’s-eye view. They organize more than 60 papers into geometry-based view transformers and cross-attention vision transformers. On nuScenes camera-only test data, BEVPoolv2 reports 0.586 mAP and BEVFormerV2 reports 0.580. These are reported results from different methods, not a controlled comparison.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.