CVMay 23, 2025

PoseBH: Prototypical Multi-Dataset Training Beyond Human Pose Estimation

arXiv:2505.17475v14 citationsh-index: 15Has CodeCVPR
Originality Highly original
AI Analysis

This work solves the problem of integrating diverse skeleton types in pose estimation for researchers and practitioners, representing an incremental advancement with novel techniques for a known bottleneck.

The paper tackles the challenge of multi-dataset training for pose estimation by addressing skeletal heterogeneity and limited supervision, resulting in improved generalization across whole-body and animal pose datasets while maintaining performance on standard benchmarks.

We study multi-dataset training (MDT) for pose estimation, where skeletal heterogeneity presents a unique challenge that existing methods have yet to address. In traditional domains, \eg regression and classification, MDT typically relies on dataset merging or multi-head supervision. However, the diversity of skeleton types and limited cross-dataset supervision complicate integration in pose estimation. To address these challenges, we introduce PoseBH, a new MDT framework that tackles keypoint heterogeneity and limited supervision through two key techniques. First, we propose nonparametric keypoint prototypes that learn within a unified embedding space, enabling seamless integration across skeleton types. Second, we develop a cross-type self-supervision mechanism that aligns keypoint predictions with keypoint embedding prototypes, providing supervision without relying on teacher-student models or additional augmentations. PoseBH substantially improves generalization across whole-body and animal pose datasets, including COCO-WholeBody, AP-10K, and APT-36K, while preserving performance on standard human pose benchmarks (COCO, MPII, and AIC). Furthermore, our learned keypoint embeddings transfer effectively to hand shape estimation (InterHand2.6M) and human body shape estimation (3DPW). The code for PoseBH is available at: https://github.com/uyoung-jeong/PoseBH.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes