AI Paper Atlas

Research map / 3D vision / Human pose estimation

Temporal Keypoint Matching and Refinement Network for Pose Estimation and Tracking

Chunluan Zhou Zhou Ren Gang Hua

ECCV 2020 · 3D vision › Human pose estimation · approach: Multi-person pose estimation · 20 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 pose-based NMS instead of conventional NMS (ablation)Temporal keypoint matching and refinement network · PoseTrack 2017 valmAP: 75.2 -> 76.1Improved
add pose-based NMS instead of conventional NMS (ablation)Temporal keypoint matching and refinement network · PoseTrack 2017 valMOTA: 63.5 -> 64.8Improved
use OKS instead of IOU similarity for tracking (ablation)Temporal keypoint matching and refinement network · PoseTrack 2017 valMOTA: 64.8 -> 65.0No clear effect
use keypoint matching (TKM) instead of IOU/OKS similarity (ablation)Temporal keypoint matching and refinement network · PoseTrack 2017 valMOTA: 64.8 -> 67.3Improved
use keypoint-level matching (TKM) instead of whole-body matching (TBM) (ablation)Temporal keypoint matching and refinement network · PoseTrack 2017 valMOTA: 66.0 -> 67.3Improved
add keypoint refinement without temporal context (ablation)Temporal keypoint matching and refinement network · PoseTrack 2017 valmAP: 76.1 -> 76.8Improved
add temporal context to keypoint refinement (ablation)Temporal keypoint matching and refinement network · PoseTrack 2017 valmAP: 76.8 -> 78.2Improved
add temporal context to keypoint refinement (ablation)Temporal keypoint matching and refinement network · PoseTrack 2017 valMOTA: 68.1 -> 69.6Improved
use ResNeXt-101 detector instead of ResNet-101 (ablation)Temporal keypoint matching and refinement network · PoseTrack 2017 valMOTA: 69.6 -> 71.4Improved
use HRNet backbone instead of ResNet-152 (ablation)Temporal keypoint matching and refinement network · PoseTrack 2017 valmAP: 79.0 -> 79.5Improved
add pose-based NMS instead of conventional NMS (ablation)Temporal keypoint matching and refinement network · PoseTrack 2018 valMOTA: 63.7 -> 65.3Improved
use keypoint matching (TKM) instead of IOU similarity (ablation)Temporal keypoint matching and refinement network · PoseTrack 2018 valMOTA: 65.3 -> 67.0Improved

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