AI Paper Atlas

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

HyperDet3D: Learning a Scene-Conditioned 3D Object Detector

Yu Zheng, Yueqi Duan, Jiwen Lu, Jie Zhou, Qi Tian

CVPR 2022 · Detection and segmentation › 3D detection and driving · approach: Unified LiDAR 3D detection · 38 citations (OpenAlex)

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 scene-conditioned hypernetworks with MSA and DDHGF3D-level 3D detection on ScanNet V2 · ScanNet V2 validationmAP@0.5: 52.8 -> 57.2Improved
add scene-conditioned hypernetworks with MSA and DDHGF3D-level 3D detection on SUN RGB-D V1 · SUN RGB-D V1 validationmAP@0.5: 45.2 -> 47.3Improved
add scene-conditioned knowledge using SSA (ablation)DDH-only baseline · ScanNet V2 validationmAP@0.5: 53.5 -> 54.7Improved
replace SSA with multi-head MSA (ablation)HyperDet3D with scene-agnostic and scene-specific knowledge · ScanNet V2 validationmAP@0.5: +2.5%Improved
use scene-specific knowledge without scene-agnostic knowledge (ablation)HyperDet3D · ScanNet V2 validationmAP@0.5: 57.2 -> 55.4Hurt
use scene-agnostic knowledge without scene-specific knowledge (ablation)HyperDet3D · ScanNet V2 validationmAP@0.5: 57.2 -> 53.8Hurt
add the disentangled detection head (ablation)MSA with scene-agnostic and scene-specific knowledge · ScanNet V2 validationmAP@0.5: 56.8 -> 57.2Improved
add a scene-type classification branchGF3D · ScanNet V2 validationmAP@0.5: 52.8 -> 54.3Improved
use HyperDet3D rather than GF3Dcross-dataset detector pretrained on SUN RGB-D V1 · Cross-dataset ScanNet V2 validationmAP8: 35.2 -> 43.6Improved
remove scene-specific knowledge (ablation)HyperDet3D in cross-dataset evaluation · Cross-dataset ScanNet V2 validationmAP8: -3.0%Hurt
replace MSA with SSA (ablation)HyperDet3D in cross-dataset evaluation · Cross-dataset ScanNet V2 validationmAP8: -3.4%Hurt
add scene-conditioned detectionGF3D on SUN RGB-D V1 bathtub detections · SUN RGB-D V1 validation, bathtubAP: 64.0 -> 59.1Hurt

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