AI Paper Atlas

Research map / 3D vision / Human pose estimation

Non-Line-of-Sight Estimation of Fast Human Motion with Slow Scanning Imagers

Javier Grau Chopite, Patrick Hähn, Matthias B Hullin

ECCV 2024 · 3D vision › Human pose estimation · approach: Physics and sensor-based motion capture · 3 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
predict center pose plus velocity with a motion-aware network instead of a static network (ablation)NLoS human pose network · AMASS random sampleMPJPE (m): 0.059 -> 0.042Improved
motion-aware network instead of static network on average velocity error (ablation)NLoS human pose network · AMASS random sampleVerr: 0.136 -> 0.129No clear effect
motion-aware network instead of static network on fast-motion failure cases (ablation)NLoS human pose network · Dance sequencePCK (%): 2.27 -> 84.09Improved
motion-aware network on a slow jog sequence (ablation)NLoS human pose network · Jog sequencePCK (%): 96.30 -> 98.57No clear effect
use a learned scan-aware model instead of Light Cone Transform filteringNLoS reconstruction of fast motion · Synthetic and real 4 Hz capturesqualitative reconstructionImproved
apply HiddenPose to low-resolution scanning data instead of the scan-aware methodHiddenPose (Liu et al.) · Real 4 Hz capturesqualitative pose accuracyHurt
real-world fast jump with the scan-aware pipelinescan-aware network · Real-world jump capturebone length consistencyHurt

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