Computer Graphics

University of California - Berkeley

KBody: Balanced monocular whole-body estimation


KBody is a method for fitting a low-dimensional body model to an image. It follows a predict-and-optimize approach, relying on data-driven model estimates for the constraints that will be used to solve for the body's parameters. Compared to other approaches, it introduces virtual joints to identify higher quality correspondences and disentangles the optimization between the pose and shape parameters to achieve a more balanced result in terms of pose and shape capturing capacity, as well as pixel alignment.


Nikolaos Zioulis and James F. O'Brien. "KBody: Balanced monocular whole-body estimation". In Proceedings of 6th Workshop on Computer Vision for Fashion, Art, and Design (CVFAD) at CVPR 2023, page 6, June 2023.

Supplemental Material

Talk given at CVFAD Workshop at CVPR 2023

Supplemental Comparisons