AI Paper Atlas

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

Harnessing Uncertainty-aware Bounding Boxes for Unsupervised 3D Object Detection

Ruiyang Zhang, Hu Zhang, Zhedong Zheng

ICCV 2025 · Detection and segmentation › 3D detection and driving · approach: Depth-based and label-free 3D detection · 11 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 UA3D uncertainty estimation and regularization to self-training (vs OYSTER)UA3D · nuScenesAP_BEV@0.25 (0-80m): 12.7 -> 19.6Improved
add UA3D (vs MODEST)UA3D · LyftAP_BEV@0.25 (0-80m): 57.3 -> 61.4Improved
add UA3D on long range (50-80m)UA3D · nuScenesAP_BEV@0.25 (50-80m): 1.7 -> 4.3Improved
add UA3D on long range (50-80m)UA3D · LyftAP_BEV@0.25 (50-80m): 27.0 -> 36.8Improved
use learnable uncertainty instead of distance-based rule uncertainty (ablation)UA3D · nuScenesAP_BEV@0.25 (0-80m): 14.8 -> 19.6Improved
use learnable uncertainty instead of num-points-based rule uncertainty (ablation)UA3D · nuScenesAP_BEV@0.25 (0-80m): 13.7 -> 19.6Improved
use learnable uncertainty instead of volume-based rule uncertainty (ablation)UA3D · nuScenesAP_BEV@0.25 (0-80m): 12.3 -> 19.6Improved
use coordinate-level instead of box-level uncertainty (ablation)UA3D · nuScenesAP_BEV@0.25 (0-80m): 17.2 -> 19.6Improved
use coordinate-level instead of point-cloud-level uncertainty (ablation)UA3D · nuScenesAP_BEV@0.25 (0-80m): 12.1 -> 19.6Improved
set auxiliary detector channel coefficient gamma=0.5 instead of 0.25 (ablation)UA3D · nuScenesAP_BEV@0.25 (0-80m): 16.9 -> 19.6Improved
set gamma=0.5 instead of 1 (same-capacity auxiliary) (ablation)UA3D · nuScenesAP_BEV@0.25 (0-80m): 14.7 -> 19.6Improved
set lambda=1e-5 instead of 1e-4 (ablation)UA3D · nuScenesAP_BEV@0.25 (0-80m): 15.2 -> 19.6Improved

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