atlas

Research map / 3D vision / Human pose estimation

Scene-Aware Generative Network for Human Motion Synthesis

CVPR 2021 · 3D vision › Human pose estimation · approach: Human motion prediction · 64 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
separately generate trajectory and pose (ablation)CSGN · GTA-IMMotion FID (Ave): 118.6 -> 32.1Helped
add depth supervision (ablation)M-only scene-aware generator · GTA-IMMotion FID (Ave): 33.1 -> 32.1Helped
add depth supervision (ablation)M-only scene-aware generator · GTA-IMNon-Collision Score (Ave): 0.891 -> 0.902Helped
add geometry-aware discriminators (ablation)CSGN + M · GTA-IMMotion FID (Ave): 32.1 -> 27.6Helped
add geometry-aware discriminators (ablation)CSGN + M · PROXMotion FID (Ave): 58.7 -> 54.6Helped
add geometry-aware discriminators (ablation)scene-aware generator · PROXNon-Collision Score (Ave): 0.742 -> 0.792Helped
add context discriminator (ablation)M + depth supervision · GTA-IMMotion FID (Ave): 32.1 -> 30.8Helped
add projection discriminator (ablation)M + depth supervision · GTA-IMMotion FID (Ave): 32.1 -> 28.7Helped
add projection and context discriminators (ablation)M + depth supervision · GTA-IMNon-Collision Score (Ave): 0.902 -> 0.943Helped

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