PoseBench3D: A Cross-Dataset Analysis Framework for 3D Human Pose Estimation via Pose Lifting Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Manzur, Saad, Vela, Bryan, Vela, Brandon, Agrawal, Aditya, Dang-Vu, Lan-Anh, Li, David, Hayes, Wayne
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914048076939264
author Manzur, Saad
Vela, Bryan
Vela, Brandon
Agrawal, Aditya
Dang-Vu, Lan-Anh
Li, David
Hayes, Wayne
author_facet Manzur, Saad
Vela, Bryan
Vela, Brandon
Agrawal, Aditya
Dang-Vu, Lan-Anh
Li, David
Hayes, Wayne
contents Reliable three-dimensional human pose estimation (3D HPE) remains challenging due to the differences in viewpoints, environments, and camera conventions among datasets. As a result, methods that achieve near-optimal in-dataset accuracy often degrade on unseen datasets. In practice, however, systems must adapt to diverse viewpoints, environments, and camera setups--conditions that differ significantly from those encountered during training, which is often the case in real-world scenarios. Measuring cross-dataset performance is a vital process, but extremely labor-intensive when done manually for human pose estimation. To address these challenges, we automate this evaluation using PoseBench3D, a standardized testing framework that enables consistent and fair cross-dataset comparisons on previously unseen data. PoseBench3D streamlines testing across four widely used 3D HPE datasets via a single, configurable interface. Using this framework, we re-evaluate 18 methods and report over 100 cross-dataset results under Protocol 1: MPJPE and Protocol 2: PA-MPJPE, revealing systematic generalization gaps and the impact of common preprocessing and dataset setup choices. The PoseBench3D code is found at: https://github.com/bryanjvela/PoseBench3D
format Preprint
id arxiv_https___arxiv_org_abs_2505_10888
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PoseBench3D: A Cross-Dataset Analysis Framework for 3D Human Pose Estimation via Pose Lifting Networks
Manzur, Saad
Vela, Bryan
Vela, Brandon
Agrawal, Aditya
Dang-Vu, Lan-Anh
Li, David
Hayes, Wayne
Computer Vision and Pattern Recognition
Reliable three-dimensional human pose estimation (3D HPE) remains challenging due to the differences in viewpoints, environments, and camera conventions among datasets. As a result, methods that achieve near-optimal in-dataset accuracy often degrade on unseen datasets. In practice, however, systems must adapt to diverse viewpoints, environments, and camera setups--conditions that differ significantly from those encountered during training, which is often the case in real-world scenarios. Measuring cross-dataset performance is a vital process, but extremely labor-intensive when done manually for human pose estimation. To address these challenges, we automate this evaluation using PoseBench3D, a standardized testing framework that enables consistent and fair cross-dataset comparisons on previously unseen data. PoseBench3D streamlines testing across four widely used 3D HPE datasets via a single, configurable interface. Using this framework, we re-evaluate 18 methods and report over 100 cross-dataset results under Protocol 1: MPJPE and Protocol 2: PA-MPJPE, revealing systematic generalization gaps and the impact of common preprocessing and dataset setup choices. The PoseBench3D code is found at: https://github.com/bryanjvela/PoseBench3D
title PoseBench3D: A Cross-Dataset Analysis Framework for 3D Human Pose Estimation via Pose Lifting Networks
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.10888