BinaryHPE: 3D Human Pose and Shape Estimation via Binarization

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Zhiteng, Zhang, Yulun, Lin, Jing, Qin, Haotong, Gu, Jinjin, Yuan, Xin, Kong, Linghe, Yang, Xiaokang
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916592749641728
author Li, Zhiteng
Zhang, Yulun
Lin, Jing
Qin, Haotong
Gu, Jinjin
Yuan, Xin
Kong, Linghe
Yang, Xiaokang
author_facet Li, Zhiteng
Zhang, Yulun
Lin, Jing
Qin, Haotong
Gu, Jinjin
Yuan, Xin
Kong, Linghe
Yang, Xiaokang
contents 3D human pose and shape estimation (HPE) aims to reconstruct the 3D human body, face, and hands from a single image. Although powerful deep learning models have achieved accurate estimation in this task, they require enormous memory and computational resources. Consequently, these methods can hardly be deployed on resource-limited edge devices. In this work, we propose BinaryHPE, a novel binarization method designed to estimate the 3D human body, face, and hands parameters efficiently. Specifically, we propose a novel binary backbone called Binarized Dual Residual Network (BiDRN), designed to retain as much full-precision information as possible. Furthermore, we propose the Binarized BoxNet, an efficient sub-network for predicting face and hands bounding boxes, which further reduces model redundancy. Comprehensive quantitative and qualitative experiments demonstrate the effectiveness of BinaryHPE, which has a significant improvement over state-of-the-art binarization algorithms. Moreover, our BinaryHPE achieves comparable performance with the full-precision method Hand4Whole while using only 22.1% parameters and 14.8% operations. We will release all the code and pretrained models.
format Preprint
id arxiv_https___arxiv_org_abs_2311_14323
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle BinaryHPE: 3D Human Pose and Shape Estimation via Binarization
Li, Zhiteng
Zhang, Yulun
Lin, Jing
Qin, Haotong
Gu, Jinjin
Yuan, Xin
Kong, Linghe
Yang, Xiaokang
Computer Vision and Pattern Recognition
3D human pose and shape estimation (HPE) aims to reconstruct the 3D human body, face, and hands from a single image. Although powerful deep learning models have achieved accurate estimation in this task, they require enormous memory and computational resources. Consequently, these methods can hardly be deployed on resource-limited edge devices. In this work, we propose BinaryHPE, a novel binarization method designed to estimate the 3D human body, face, and hands parameters efficiently. Specifically, we propose a novel binary backbone called Binarized Dual Residual Network (BiDRN), designed to retain as much full-precision information as possible. Furthermore, we propose the Binarized BoxNet, an efficient sub-network for predicting face and hands bounding boxes, which further reduces model redundancy. Comprehensive quantitative and qualitative experiments demonstrate the effectiveness of BinaryHPE, which has a significant improvement over state-of-the-art binarization algorithms. Moreover, our BinaryHPE achieves comparable performance with the full-precision method Hand4Whole while using only 22.1% parameters and 14.8% operations. We will release all the code and pretrained models.
title BinaryHPE: 3D Human Pose and Shape Estimation via Binarization
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.14323