arXiv cs.AI· Xuyi Hu, Jin Lyu, Jiuming Liu, Yebin Liu, Silvia Zuffi, Liang An, Stefan Goetz·· 9 天前AI 评分22
SAM 3D Animal：从野外图像进行可提示的动物 3D 重建
SAM 3D Animal: Promptable Animal 3D Reconstruction from Images in the Wild
AI 导读
研究者提出 SAM 3D Animal，首个可从单张图像对多只动物进行 3D 重建的可提示框架，基于 SMAL+ 参数化动物模型，支持关键点和掩码提示以应对拥挤遮挡场景。团队同时发布含超 5K 图像的多动物 3D 数据集 Herd3D，该方法在 Animal3D、APTv2 和 Animal Kingdom 上取得 SOTA，论文入选 NeurIPS 2026 Oral。
正文 · 原文
Abstract:3D animal reconstruction in the wild remains challenging due to large species variation, frequent occlusions, and the prevalence of multi-animal scenes, while existing methods predominantly focus on single-animal settings. We present SAM 3D Animal, the first promptable framework for multi-animal 3D reconstruction from a single image. Built on the SMAL+ parametric animal model, our method jointly reconstructs multiple instances and supports flexible prompts in the form of keypoints and masks which enable more reliable disambiguation in crowded and occluded scenes. To train such a model, we further introduce Herd3D, a multi-animal 3D dataset containing over 5K images, designed to increase diversity in species, interactions, and occlusion patterns. Experiments on the Animal3D, APTv2, and Animal Kingdom datasets show that our framework achieves state-of-the-art results over both existing model-based and model-free methods, demonstrating a scalable and effective solution for prompt-driven animal 3D reconstruction in the wild.
| Comments: | NeurIPS 2026 Oral. Project website: this https URL |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.07604 [cs.CV] |
| (or arXiv:2605.07604v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2605.07604 arXiv-issued DOI via DataCite |
Submission history
From: Xuyi Hu [view email]
[v1]
Fri, 8 May 2026 11:26:46 UTC (7,493 KB)
[v2]
Thu, 8 Oct 2026 12:53:00 UTC (26,118 KB)
来源：arXiv cs.AI · arxiv.org