English
Russian
French
Spanish
German
Japanese
korean
PortugueseNews
Oct 22, 2025
The Importance of Disentanglement: SAGE Outperforms Unified VQ-VAE Baselines in...
This article presents an ablation study confirming that disentangling motion latents into upper and lower halves significantly enh...
Oct 22, 2025
Quantitative and Qualitative Results: SAGE Outperforms SOTA in Full-Body 3D Avat...
This article presents the quantitative and qualitative results for the SAGE model across three evaluation settings
Oct 22, 2025
Quantifying Motion Accuracy: MPJRE, MPJPE, and Consistency Metrics for HMD-based...
This article outlines the evaluation metrics used to benchmark 3D avatar reconstruction models on the AMASS motion capture dataset...
Oct 22, 2025
Temporal Refinement in Stratified Motion Diffusion: Utilizing GRU for Smoothed F...
This article describes the online inference stage of the model, which predicts 3D avatars in a frame-by-frame manner using a slidi...
Oct 21, 2025
SAGE: Stratified Avatar Generation for Full-Body 3D Reconstruction from Sparse H...
This paper introduces Stratified Avatar Generation, a novel, decoupled two-stage approach for reconstructing 3D full-body avatars...
