AI Paper Atlas

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

When the City Teaches the Car: Label-Free 3D Perception from Infrastructure

ZHEN XU, Jinsu Yoo, Cristian Bautista, Zanming Huang, Tai-Yu Pan, Zhenzhen Liu, Katie Luo, Mark Campbell, Bharath Hariharan, Wei-Lun Chao

ECCV 2026 · Detection and segmentation › 3D detection and driving · approach: Depth-based and label-free 3D detection · 0 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 tracking-based box refinement to PP-score pseudo-labels (ablation)RSU pseudo-labels, CenterPoint ego · Town10 ego testaverage AP: 72.7 -> 76.7Improved
add tracking-based refinement (ablation)RSU pseudo-labels, PointPillars ego · Town10 ego testaverage AP: 73.1 -> 78.6Improved
simulate communication noise in RSU-to-ego broadcasting (ablation)PP score + tracking, CenterPoint · Town10 ego testaverage AP: 76.7 -> 66.4Hurt
add heuristic box refinement under communication noise (ablation)PP score + tracking, CenterPoint · Town10 ego testaverage AP: 66.4 -> 69.8Improved
add heuristic box refinement under communication noise (ablation)PP score + tracking, CenterPoint · Town10 ego test (Car AP)Car AP: 80.4 -> 77.7Hurt
remove box filtering of distant or sparse boxes (ablation)pseudo-label aggregation · Town10 ego testaverage AP: 60.0 -> 66.4Improved
use supervised RSU detectors instead of unsupervisedCenterPoint ego · Town10 ego testaverage AP: 76.7 -> 90.6Improved
scale to 4 towns and 48 RSUsCenterPoint ego · combined 4-town testaverage AP: 80.7 vs 76.7 (single town)No clear effect
fine-tune rural-trained ego detector with urban infra pseudo-labelsCenterPoint ego · Town10 urban testaverage AP: 64.3 -> 79.3Improved
combine infra pseudo-labels with Oyster pseudo-labelsPointRCNN · dynamic-object APDyn AP: 62.0 -> 64.9Improved
combine infra pseudo-labels with Oyster pseudo-labels under comm. noisePointRCNN · dynamic-object APDyn AP: 49.9 -> 59.5Improved
increase RSUs from 4 or 8 to 12ego detector · Town10 ego testAPImproved

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