PEAR: Pixel-aligned Expressive humAn mesh Recovery

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Jiahao, Liu, Yunfei, Lin, Lijian, Zhu, Ye, Zhu, Lei, Li, Jingyi, Li, Yu
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915777561493504
author Wu, Jiahao
Liu, Yunfei
Lin, Lijian
Zhu, Ye
Zhu, Lei
Li, Jingyi
Li, Yu
author_facet Wu, Jiahao
Liu, Yunfei
Lin, Lijian
Zhu, Ye
Zhu, Lei
Li, Jingyi
Li, Yu
contents Reconstructing detailed 3D human meshes from a single in-the-wild image remains a fundamental challenge in computer vision. Existing SMPLX-based methods often suffer from slow inference, produce only coarse body poses, and exhibit misalignments or unnatural artifacts in fine-grained regions such as the face and hands. These issues make current approaches difficult to apply to downstream tasks. To address these challenges, we propose PEAR-a fast and robust framework for pixel-aligned expressive human mesh recovery. PEAR explicitly tackles three major limitations of existing methods: slow inference, inaccurate localization of fine-grained human pose details, and insufficient facial expression capture. Specifically, to enable real-time SMPLX parameter inference, we depart from prior designs that rely on high resolution inputs or multi-branch architectures. Instead, we adopt a clean and unified ViT-based model capable of recovering coarse 3D human geometry. To compensate for the loss of fine-grained details caused by this simplified architecture, we introduce pixel-level supervision to optimize the geometry, significantly improving the reconstruction accuracy of fine-grained human details. To make this approach practical, we further propose a modular data annotation strategy that enriches the training data and enhances the robustness of the model. Overall, PEAR is a preprocessing-free framework that can simultaneously infer EHM-s (SMPLX and scaled-FLAME) parameters at over 100 FPS. Extensive experiments on multiple benchmark datasets demonstrate that our method achieves substantial improvements in pose estimation accuracy compared to previous SMPLX-based approaches. Project page: https://wujh2001.github.io/PEAR
format Preprint
id arxiv_https___arxiv_org_abs_2601_22693
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PEAR: Pixel-aligned Expressive humAn mesh Recovery
Wu, Jiahao
Liu, Yunfei
Lin, Lijian
Zhu, Ye
Zhu, Lei
Li, Jingyi
Li, Yu
Computer Vision and Pattern Recognition
Artificial Intelligence
Reconstructing detailed 3D human meshes from a single in-the-wild image remains a fundamental challenge in computer vision. Existing SMPLX-based methods often suffer from slow inference, produce only coarse body poses, and exhibit misalignments or unnatural artifacts in fine-grained regions such as the face and hands. These issues make current approaches difficult to apply to downstream tasks. To address these challenges, we propose PEAR-a fast and robust framework for pixel-aligned expressive human mesh recovery. PEAR explicitly tackles three major limitations of existing methods: slow inference, inaccurate localization of fine-grained human pose details, and insufficient facial expression capture. Specifically, to enable real-time SMPLX parameter inference, we depart from prior designs that rely on high resolution inputs or multi-branch architectures. Instead, we adopt a clean and unified ViT-based model capable of recovering coarse 3D human geometry. To compensate for the loss of fine-grained details caused by this simplified architecture, we introduce pixel-level supervision to optimize the geometry, significantly improving the reconstruction accuracy of fine-grained human details. To make this approach practical, we further propose a modular data annotation strategy that enriches the training data and enhances the robustness of the model. Overall, PEAR is a preprocessing-free framework that can simultaneously infer EHM-s (SMPLX and scaled-FLAME) parameters at over 100 FPS. Extensive experiments on multiple benchmark datasets demonstrate that our method achieves substantial improvements in pose estimation accuracy compared to previous SMPLX-based approaches. Project page: https://wujh2001.github.io/PEAR
title PEAR: Pixel-aligned Expressive humAn mesh Recovery
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2601.22693