Skip to content

Opening book details…

About this document

You Co-Evolution of Pose and Mesh For 3D Human Body Estimation ICCV 2023 Paper by sina mazarei is a document available to read on EtoBox.

The paper presents a Pose and Mesh Co-Evolution network (PMCE) designed to improve 3D human body estimation from video by decoupling the task into video-based 3D pose estimation and mesh vertices regression. PMCE utilizes a two-stream encoder to estimate mid-frame 3D poses and extract temporal image features, followed by a co-evolution decoder that enhances pose and mesh interactions using Adaptive Layer Normalization. Experimental results demonstrate that PMCE outperforms existing methods in accuracy and t

Author
sina mazarei
Language
EN