Cross-view and Cross-pose Completion for 3D Human Understanding

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Armando, Matthieu, Galaaoui, Salma, Baradel, Fabien, Lucas, Thomas, Leroy, Vincent, Brégier, Romain, Weinzaepfel, Philippe, Rogez, Grégory
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909173159034880
author Armando, Matthieu
Galaaoui, Salma
Baradel, Fabien
Lucas, Thomas
Leroy, Vincent
Brégier, Romain
Weinzaepfel, Philippe
Rogez, Grégory
author_facet Armando, Matthieu
Galaaoui, Salma
Baradel, Fabien
Lucas, Thomas
Leroy, Vincent
Brégier, Romain
Weinzaepfel, Philippe
Rogez, Grégory
contents Human perception and understanding is a major domain of computer vision which, like many other vision subdomains recently, stands to gain from the use of large models pre-trained on large datasets. We hypothesize that the most common pre-training strategy of relying on general purpose, object-centric image datasets such as ImageNet, is limited by an important domain shift. On the other hand, collecting domain-specific ground truth such as 2D or 3D labels does not scale well. Therefore, we propose a pre-training approach based on self-supervised learning that works on human-centric data using only images. Our method uses pairs of images of humans: the first is partially masked and the model is trained to reconstruct the masked parts given the visible ones and a second image. It relies on both stereoscopic (cross-view) pairs, and temporal (cross-pose) pairs taken from videos, in order to learn priors about 3D as well as human motion. We pre-train a model for body-centric tasks and one for hand-centric tasks. With a generic transformer architecture, these models outperform existing self-supervised pre-training methods on a wide set of human-centric downstream tasks, and obtain state-of-the-art performance for instance when fine-tuning for model-based and model-free human mesh recovery.
format Preprint
id arxiv_https___arxiv_org_abs_2311_09104
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Cross-view and Cross-pose Completion for 3D Human Understanding
Armando, Matthieu
Galaaoui, Salma
Baradel, Fabien
Lucas, Thomas
Leroy, Vincent
Brégier, Romain
Weinzaepfel, Philippe
Rogez, Grégory
Computer Vision and Pattern Recognition
Human perception and understanding is a major domain of computer vision which, like many other vision subdomains recently, stands to gain from the use of large models pre-trained on large datasets. We hypothesize that the most common pre-training strategy of relying on general purpose, object-centric image datasets such as ImageNet, is limited by an important domain shift. On the other hand, collecting domain-specific ground truth such as 2D or 3D labels does not scale well. Therefore, we propose a pre-training approach based on self-supervised learning that works on human-centric data using only images. Our method uses pairs of images of humans: the first is partially masked and the model is trained to reconstruct the masked parts given the visible ones and a second image. It relies on both stereoscopic (cross-view) pairs, and temporal (cross-pose) pairs taken from videos, in order to learn priors about 3D as well as human motion. We pre-train a model for body-centric tasks and one for hand-centric tasks. With a generic transformer architecture, these models outperform existing self-supervised pre-training methods on a wide set of human-centric downstream tasks, and obtain state-of-the-art performance for instance when fine-tuning for model-based and model-free human mesh recovery.
title Cross-view and Cross-pose Completion for 3D Human Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.09104