AniMer: Animal Pose and Shape Estimation Using Transformer

Jin Lyu, Tianyi Zhu, Yi Gu, Li Lin, Pujin Cheng, Yebin Liu, Xiaoying Tang, Liang An

Keywords: Animal Pose and Shape Estimation; Transformer; Synthetic Dataset

Abstract: Quantitative analysis of animal behavior and biomechanics requires accurate animal pose and shape estimation across species, and is important for animal welfare and biological research. However, the small network capacity of previous methods and limited multi-species dataset leave this problem underexplored. To this end, this paper presents AniMer to estimate animal pose and shape using Transformer, enhancing the reconstruction accuracy of diverse quadrupedal species. AniMer aims to unify the understanding of various quadrupedal forms within a single framework, overcoming the limitations of traditional methods that focus on narrow specific species. A key feature of AniMer is its integration of a high-capacity Transformer-based backbone, which significantly boosts performance. To effectively train AniMer, we aggregate most available open-source quadrupedal datasets, either with 3D or 2D labels, and introduce CtrlAni3D, a novel large-scale synthetic dataset created through a diffusion-based image generation model, consisting of 9.7k pixel-aligned SMAL mesh-labeled images. This combination of a robust backbone and an expansive dataset enables AniMer to outperform existing methods on the multi-species Animal3D dataset and singlespecies dog benchmarks. Experiments on the unseen AnimalKingdom dataset further demonstrate the effectiveness of CtrlAni3D in enhancing generalization capabilities. Our study, through the development of AniMer and CtrlAni3D, underscores the significance of a large-capacity backbone and AI-driven synthetic data generation in advancing animal pose estimation research. Code and data will be released upon publication.

Primary Area: applications to computer vision, audio, language, and other modalities

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 1845