AI Paper Atlas

Research map / 3D vision / 3D generation

SignAvatars: A Large-scale 3D Sign Language Holistic Motion Dataset and Benchmark

Zhengdi Yu, Shaoli Huang, yongkang cheng, Tolga Birdal

ECCV 2024 · 3D vision › 3D generation · approach: Scene-aware human motion synthesis · 39 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 biomechanical hand constraints to the fitting objective (ablation)SignAvatars annotation pipeline · EHFPA-MPVPE (Hands): 5.4 -> 4.7Improved
add biomechanical hand constraints to the fitting objective (ablation)SignAvatars annotation pipeline · EHFMPVPE (Hands): 12.5 -> 9.7Improved
add biomechanical hand constraints to the fitting objective (ablation)SignAvatars annotation pipeline · EHFPA-MPJPE (Body): 16.5 -> 15.6Improved
use the optimization-based annotation pipeline instead of regression-based holistic methods3D holistic mesh recovery · EHFPA-MPVPE (Holistic): 48.7 -> 12.9 (vs OSX)Improved
replace pre-trained CLIP features with the parallel linguistic feature generator (PLFG) (ablation)SignVAE · SignAvatars HamNoSys holistic subsetR-Precision top-1: 0.385 -> 0.429Improved
replace pre-trained CLIP features with PLFG (ablation)SignVAE · SignAvatars HamNoSys holistic subsetMM-dist: 3.056 -> 2.651Improved
use VQVAE backbone instead of diffusion (MDM) with the same linguistic features (ablation)SignVAE vs SignDiffuse · SignAvatars HamNoSys holistic subsetR-Precision top-1: 0.285 -> 0.429Improved
use 3D holistic mesh-derived motion instead of 2D cues for HamNoSys production (ablation)Ham2Pose · SignAvatars HamNoSys holistic subsetDTW-MJE rank top-1: 0.092 -> 0.253Improved
evaluate arms only (Gesture) instead of holistic motion (ablation)SignVAE · SignAvatars HamNoSys subsetMM-dist: FID 0.884 -> 0.851; MM-dist 2.651 -> 2.876Mixed
use word-level prompts instead of spoken-language promptsSignVAE · SignAvatars holisticFID: 4.359 -> 0.756Improved
train and evaluate on mesh-derived 3D keypoints instead of 3D-lifted OpenPose keypoints (ablation)Progressive Transformer on Phoenix-2014T · Phoenix-2014T (eval model C_ours)R-Precision top-1: 0.095 -> 0.134Improved

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