KBody: Towards general, robust, and aligned monocular whole-body estimation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zioulis, Nikolaos, O'Brien, James F.
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909224211054592
author Zioulis, Nikolaos
O'Brien, James F.
author_facet Zioulis, Nikolaos
O'Brien, James F.
contents 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. Acknowledging the importance of high quality correspondences, it leverages ``virtual joints" to improve fitting performance, disentangles the optimization between the pose and shape parameters, and integrates asymmetric distance fields to strike a balance in terms of pose and shape capturing capacity, as well as pixel alignment. We also show that generative model inversion offers a strong appearance prior that can be used to complete partial human images and used as a building block for generalized and robust monocular body fitting. Project page: https://zokin.github.io/KBody.
format Preprint
id arxiv_https___arxiv_org_abs_2304_11542
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle KBody: Towards general, robust, and aligned monocular whole-body estimation
Zioulis, Nikolaos
O'Brien, James F.
Computer Vision and Pattern Recognition
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. Acknowledging the importance of high quality correspondences, it leverages ``virtual joints" to improve fitting performance, disentangles the optimization between the pose and shape parameters, and integrates asymmetric distance fields to strike a balance in terms of pose and shape capturing capacity, as well as pixel alignment. We also show that generative model inversion offers a strong appearance prior that can be used to complete partial human images and used as a building block for generalized and robust monocular body fitting. Project page: https://zokin.github.io/KBody.
title KBody: Towards general, robust, and aligned monocular whole-body estimation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2304.11542