AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Better Regression Makes Better Test-time Adaptive 3D Object Detection

Jiakang Yuan, Bo Zhang, Kaixiong Gong, Xiangyu Yue, Botian Shi, Yu Qiao, Tao Chen

ECCV 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Domain-adaptive object detection

Read the paper

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
adapt only the regression layers instead of BN layers (ablation)SECOND-IoU test-time adaptation · Waymo -> KITTIAP BEV / AP3D: 52.84 / 31.77 -> 59.38 / 45.54Improved
adapt regression layers instead of all parameters on nuScenes -> KITTI (ablation)SECOND-IoU · nuScenes -> KITTIAP BEV / AP3D: 33.23 / 19.60 -> 33.63 / 21.80Improved
apply existing TTA methods (TENT, SHOT, MATE) to 3D detectionSECOND-IoU · Waymo -> KITTIAP BEV / AP3D: 46.64 / 28.83 -> 31.34 / 19.47 (TENT)Hurt
add noise-consistency pseudo-label generation (NPG) (ablation)naive pseudo-labels · Waymo -> KITTImean AP3D: 42.93 -> 44.91Improved
add confidence-guided regression refinement (CRR) (ablation)naive pseudo-labels · Waymo -> KITTImean AP3D: 42.93 -> 45.61Improved
add class-balanced EMA updating (CBU) (ablation)naive pseudo-labels + NPG + CRR · Waymo -> KITTImean AP3D: 47.97 -> 49.96Improved
use naive pseudo-labels without NPG/CRR/CBU (ablation)source model · Waymo -> KITTImean AP3D: 42.01 -> 42.93Improved
vary noise inconsistency threshold tau and size threshold p (ablation)Reg-TTA3D · Waymo -> KITTIAP BEV / AP3D: tau 1.0/1.5/2.0: 49.12 / 49.96 / 48.88 AP3DNo clear effect
apply Reg-TTA3D in cross-weather settingsCenterPoint · nuScenes -> nuScenes-C SnowmAP: 45.86 -> 47.09Improved
Reg-TTA3D vs UDA on cross-beam settingsReg-TTA3D · nuScenes -> KITTI, Waymo -> nuScenesAP3D: nuScenes->KITTI 23.13 vs ReDB 26.11Mixed

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