AI Paper Atlas

Research map / 3D vision / Human pose estimation

DETRPose: Real-Time End-to-End Multi-Person Pose Estimation via Modified Transformer Decoder and Novel Denoising Keypoints

Sebastian Janampa, Marios Pattichis

ECCV 2026 · 3D vision › Human pose estimation · approach: Multi-person pose estimation · 7 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 denoising keypoints (DK) alone (ablation)DETRPose-L, 60 queries · COCO valAP: 68.9 -> 70.0Improved
add KSVF loss alone (ablation)DETRPose-L, 60 queries · COCO valAP: 68.9 -> 69.9Improved
add Pose-LQE alone (ablation)DETRPose-L, 60 queries · COCO valAP: 68.9 -> 69.5Improved
add all of KSVF, Pose-LQE and denoising keypoints (ablation)DETRPose-L, 60 queries · COCO valAP: 68.9 -> 71.5Improved
add denoising keypoints on top of Pose-LQE (ablation)DETRPose-L, 60 queries · COCO valAP: 69.5 -> 69.6Mixed
add denoising keypoints on top of KSVF (ablation)DETRPose-L, 60 queries · COCO valAP: 69.9 -> 70.9Improved
reduce instance queries from 100 to 60 (ablation)RT-GroupPose (DETRPose-L) · COCO valAP: 69.9 -> 68.9Hurt
initialize with D-FINE COCO-pretrained weights instead of none (ablation)DETRPose-L · COCO valAP: 71.5 -> 71.1Hurt
initialize with D-FINE Object365 weights instead of none (ablation)DETRPose-L · COCO valAP: 71.5 -> 72.7Improved
use official COCO kappa values instead of constant kappa (box area) (ablation)denoising keypoints, DETRPose-L · COCO valAP: 71.1 -> 71.2Improved
compute area from masks instead of box area (constant kappa) (ablation)denoising keypoints, DETRPose-L · COCO valAP: 71.1 -> 71.3Improved
use both COCO kappa values and mask area (ablation)denoising keypoints, DETRPose-L · COCO valAP: 71.1 -> 71.5Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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