arXiv Analytics

Sign in

arXiv:2205.06254 [cs.CV]AbstractReferencesReviewsResources

Learned Vertex Descent: A New Direction for 3D Human Model Fitting

Enric Corona, Gerard Pons-Moll, Guillem Alenyà, Francesc Moreno-Noguer

Published 2022-05-12Version 1

We propose a novel optimization-based paradigm for 3D human model fitting on images and scans. In contrast to existing approaches that directly regress the parameters of a low-dimensional statistical body model (e.g. SMPL) from input images, we train an ensemble of per-vertex neural fields network. The network predicts, in a distributed manner, the vertex descent direction towards the ground truth, based on neural features extracted at the current vertex projection. At inference, we employ this network, dubbed LVD, within a gradient-descent optimization pipeline until its convergence, which typically occurs in a fraction of a second even when initializing all vertices into a single point. An exhaustive evaluation demonstrates that our approach is able to capture the underlying body of clothed people with very different body shapes, achieving a significant improvement compared to state-of-the-art. LVD is also applicable to 3D model fitting of humans and hands, for which we show a significant improvement to the SOTA with a much simpler and faster method.

Comments: Project page:
Categories: cs.CV
Related articles: Most relevant | Search more
arXiv:2104.11225 [cs.CV] (Published 2021-04-22)
Pri3D: Can 3D Priors Help 2D Representation Learning?
arXiv:1610.08851 [cs.CV] (Published 2016-10-27)
Single- and Multi-Task Architectures for Tool Presence Detection Challenge at M2CAI 2016
arXiv:1804.03867 [cs.CV] (Published 2018-04-11)
Hybrid Binary Networks: Optimizing for Accuracy, Efficiency and Memory