AI Paper Atlas

Research map / 3D vision / Human pose estimation

GTPT: Group-based Token Pruning Transformer for Efficient Human Pose Estimation

Haonan Wang, Jie Liu, Jie Tang, Gangshan Wu, Bo Xu, Yanbing Chou, Yong Wang

ECCV 2024 · 3D vision › Human pose estimation · approach: Heatmap and coordinate pose encoding · 10 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
group keypoint tokens (head, upper body, lower body) (ablation)GTPT-S · COCO-WholeBody valWhole-body AP: 58.2 -> 59.1Improved
add per-group masking of visual tokens (ablation)GTPT-S · COCO-WholeBody valWhole-body AP: 59.1 -> 59.6Improved
use MHGA with shared sparse keypoint tokens instead of MHSA within groups (ablation)GTPT-S · COCO-WholeBody valWhole-body AP: 58.8 -> 59.6Improved
introduce keypoints coarse-to-fine (human-sparse-dense) instead of all dense at once (ablation)GTPT-S · COCO-WholeBody valGFLOPs / AP: 2.26 -> 1.99 GFLOPs; 59.8 -> 59.6 APMixed
add global perceived loss to pruning (ablation)GTPT-S with grouping and masking · COCO-WholeBody valWhole-body AP: 59.2 -> 59.6Improved
prune visual tokens (ablation)GTPT-S · COCO-WholeBody valGFLOPs: 3.53 -> 1.99 (-44%)Improved
add grouping and masking when pruning directly (ablation)GTPT-S · COCO-WholeBody valWhole-body AP: 56.6 -> 59.2Improved
add curriculum learning (body then whole-body) (ablation)GTPT-S · COCO-WholeBody valWhole-body AP: 56.5 -> 59.6Improved
token pruning on single-instance latency (ablation)GTPT-S · COCO-WholeBody valFPS: 93 -> 91 FPSNo clear effect
token pruning on batch throughput and memory (ablation)GTPT-S · COCO-WholeBody valthroughput: 701 -> 1231Improved

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