no code implementations • 29 Jun 2023 • Sungho Chun, Sungbum Park, Ju Yong Chang
We show that representation learning of vertex heatmaps using an autoencoder helps improve the performance of such approaches.
Ranked #1 on 3D Human Pose Estimation on Human3.6M (using extra training data)
no code implementations • 24 Aug 2022 • Sungho Chun, Sungbum Park, Ju Yong Chang
The proposed method first estimates the coordinates of mesh vertices through a CNN-based model from input images, and acquires SMPL parameters by fitting the SMPL model to the estimated vertices.
Ranked #3 on 3D Human Pose Estimation on Human3.6M
no code implementations • 1 Dec 2021 • Seong Hyun Kim, Sunwon Jeong, Sungbum Park, Ju Yong Chang
To address this issue, this paper presents a camera motion agnostic approach for predicting 3D human pose and mesh defined in the world coordinate system.