AI Paper Atlas

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

ViewFormer: Exploring Spatiotemporal Modeling for Multi-View 3D Occupancy Perception via View-Guided Transformers

Jinke Li, Xiao He, Chonghua Zhou, Xiaoqiang Cheng, Yang Wen, Dan Zhang

ECCV 2024 · Detection and segmentation › 3D detection and driving · approach: LiDAR perception for driving · 30 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
replace projection-first deformable attention with learning-first view attention (ablation)ViewFormer · FlowOcc3D (no flow head)mIoU: 42.39 -> 43.61Improved
learn sample points in a query-specific view coordinate system instead of ego-centric coordinates (ablation)view attention · FlowOcc3D (no flow head)mIoU: 42.86 -> 43.61Improved
add streaming temporal attention (ablation)ViewFormer · FlowOcc3D (with flow head)mIoU: 39.28 -> 42.54Improved
compress voxel queries to BEV queries for temporal interaction (BEV-to-BEV vs Voxel-to-BEV) (ablation)streaming temporal attention · FlowOcc3D (with flow head)mIoU: 41.62 -> 42.54Improved
voxel-to-BEV temporal interaction flow convergence (ablation)streaming temporal attention · FlowOcc3D (with flow head)mAVE: 0.412 -> 1.104Hurt
increase streaming memory queue length from 1 to 3 frames (ablation)streaming temporal attention · FlowOcc3D (with flow head)mIoU: 41.67 -> 42.42Improved
streaming memory attention adds FPS cost (ablation)ViewFormer · FlowOcc3DFPS: 4.2 -> 4.1No clear effect
train temporal attention with streaming video instead of sliding window (ablation)ViewFormer · FlowOcc3DmIoU: 42.17 -> 42.54; 36 h -> 13 hImproved
use deformable attention instead of CNN-based temporal interaction (ablation)streaming temporal attention · FlowOcc3DmIoU: 41.95 -> 42.54Improved
apply view attention to MapTRMapTR · nuScenes HD mapmAP: 44.73 -> 50.55Improved
apply view attention to DETR3DDETR3D · nuScenes detectionmAP: 0.347 -> 0.388Improved
supervise flow with occupancy-level instead of object-level flowViewFormer · FlowOcc3D (qualitative)flow qualityImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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