AI Paper Atlas

Research map / 3D vision / Human pose estimation

Occlusion-aware Hand Pose Estimation Using Hierarchical Mixture Density Network

Qi Ye, Tae-Kyun Kim

ECCV 2018 · 3D vision › Human pose estimation · approach: Depth-based hand pose regression · 58 citations (OpenAlex)

Read the paper

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 visibility hierarchy (HMDN instead of MDN), J=20MDN · EgoBigHandVisible error (mm): 34.0 -> 30.7Improved
add visibility hierarchy (HMDN instead of MDN), J=20MDN · EgoBigHandOccluded error (mm): 36.4 -> 34.4Improved
HMDN instead of SGN, occluded jointsSGN · EgoBigHandOccluded error (mm): 36.5 -> 34.4Improved
HMDN instead of SGN, visible jointsSGN · EgoBigHandVisible error (mm): 32.8 -> 30.5Improved
mixture density (MDN) instead of single GaussianSGN · EgoBigHandOccluded min-distance error (mm): 38.9 -> 34.8Improved
HMDN instead of MDN on min-distance error to multiple labelsMDN · EgoBigHand (J=10)Occluded min-distance error (mm): 34.8 -> 34.6No clear effect
HMDN instead of MDN on min-distance error, J=30MDN · EgoBigHand (J=30)Occluded min-distance error (mm): 35.0 -> 34.5Improved
increase number of Gaussians from 10 to 30 (ablation)HMDN · EgoBigHandOccluded error (mm): 34.8 -> 34.2No clear effect
sample from visibility distribution in training (HMDN_soft vs hard), J=10 (ablation)HMDN_hard · EgoBigHandVisible error (mm): 32.2 -> 30.5Improved
sample from visibility distribution in training (soft), J=10 (ablation)HMDN_hard · EgoBigHandOccluded error (mm): 35.8 -> 34.8Improved
sample from visibility distribution in training (soft), J=20 (ablation)HMDN_hard · EgoBigHandOccluded error (mm): 35.9 -> 34.4Improved
sample from visibility distribution in training (soft), J=30 (ablation)HMDN_hard · EgoBigHandOccluded error (mm): 36.4 -> 34.2Improved

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