HUMBI is a large-scale multi-view human representation dataset. The dataset contains human representations in natural clothing. The goal of the dataset is to facilitate modeling of specific views and geometries of gaze, face, hands, body, and clothing for a specific group of people. The dataset contains 772 subjects of different genders, races, ages, and physical conditions captured by 107 synchronized high-definition cameras. Using multi-view, 3D mesh models, high-fidelity human representations are built. HUMBI is very effective in learning and building complete human models.

Citation

@InProceedings{Yu_2020_CVPR,
author = {Yu, Zhixuan and Yoon, Jae Shin and Lee, In Kyu and Venkatesh, Prashanth and Park, Jaesik and Yu, Jihun and Park, Hyun Soo},
title = {HUMBI: A Large Multiview Dataset of Human Body Expressions},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2020}
}