AI Paper Atlas

Research map / 3D vision / Human pose estimation

You Only Learn One Query: Learning Unified Human Query for Single-Stage Multi-Person Multi-Task Human-Centric Perception

Sheng Jin, Shuhuai Li, Tong Li, Wentao Liu, Chen Qian, Ping Luo

ECCV 2024 · 3D vision › Human pose estimation · approach: Multi-person pose estimation · 12 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
co-train pose with detection and segmentation (add P to D+S) (ablation)HQNet R-50 · COCO-UniHuman val detectionAP: 73.0 -> 74.5Improved
co-train pose with detection and segmentation (add P to D+S) (ablation)HQNet R-50 · COCO-UniHuman val segmentationAP: 63.6 -> 65.7Improved
add classification (gender and age) tasks to D+S+P (ablation)HQNet R-50 · COCO-UniHuman val detectionAP: 74.5 -> 74.9Improved
add classification tasks to D+S+P (ablation)HQNet R-50 · COCO-UniHuman val keypointsAP: 69.5 -> 69.3No clear effect
add Gender-aided human Model Selection (GaMS) for SMPL model choice (ablation)HQNet R-50 · COCO-UniHuman val 3D meshPA-MPJPE (lower better): 54.92 -> 50.80Improved
add GaMS (ablation)HQNet R-50 · COCO-UniHuman val 3D meshMPJPE (lower better): 87.00 -> 84.74Improved
use HQ-Ins matching (detection, pose, segmentation losses) instead of detection-only matching (ablation)HQNet · qualitativematching errorsImproved
train queries with multiple tasks instead of detection only (ablation)HQNet R-50 queries as re-ID features · PoseTrack21 MOTIDF1: 62.4 -> 64.6Improved
add auxiliary denoising losses for segmentation and pose (ablation)HQNet · COCO-UniHumanoverallNo clear effect
use ViT-L instead of R-50 backboneHQNet · COCO-UniHuman val keypointsAP: 69.3 -> 75.3Improved
single-stage unified model versus top-down ViTPose with ViT-L2D keypoint estimation · COCO-UniHuman valkeypoint AP: 78.2 vs 75.3Hurt
freeze backbone and fine-tune HQNet for new face detection taskHQNet R-50 · COCO-UniHuman val face detectionAP: 58.2 -> 68.4Improved

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