AI Paper Atlas

Research map / Detection and segmentation / 3D detection and driving

BEVFormer: Learning Bird’s-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers

Zhiqi Li, Wenhai Wang, Hongyang Li, Enze Xie, Chonghao Sima, Tong Lu, Yu Qiao, Jifeng Dai

ECCV 2022 · Detection and segmentation › 3D detection and driving · approach: Multi-camera BEV detection · 1,685 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add temporal self-attention (ablation)BEVFormer-S · nuScenes valNDS: 0.448 -> 0.517Improved
add temporal self-attention (ablation)BEVFormer-S · nuScenes valmAVE: 0.802 -> 0.394Improved
add temporal self-attention (ablation)BEVFormer-S R101 · nuScenes testNDS: 0.462 -> 0.535Improved
use local deformable spatial attention instead of global attention (ablation)BEVFormer-S · nuScenes valNDS: 0.404 -> 0.448Improved
use local deformable attention instead of reference-point-only interaction (ablation)BEVFormer-S · nuScenes valNDS: 0.423 -> 0.448Improved
increase training frames from 1 to 4 (ablation)BEVFormer · nuScenes valNDS: 0.448 -> 0.517Improved
increase training frames from 4 to 5 (ablation)BEVFormer · nuScenes valNDS: 0.517 -> 0.517No clear effect
add ego-motion alignment of history BEV features (ablation)BEVFormer · nuScenes valNDS: 51.0% NDS -> 51.7% NDSImproved
add random sampling of 4 frames from 5 (ablation)BEVFormer · nuScenes valNDS: 51.3% NDS -> 51.7% NDSImproved
use both BEV queries and history BEV features to predict offsets (ablation)BEVFormer · nuScenes valNDS: 51.3% NDS -> 51.7% NDSImproved
reduce encoder to 1 layer (config C) (ablation)BEVFormer · nuScenes valNDS: 0.517 -> 0.501Hurt
single-scale features, 100x100 BEV, 1 layer (config D) (ablation)BEVFormer · nuScenes valNDS: 0.517 -> 0.478Hurt

Later papers that compared with this method

32 later accepted papers report BEVFormer as a baseline in their results tables; 30 of them report a better number on most of the results they share with it.

24 more not listed.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy