用大模型重建单图中人与物体的3D交互，精度高且更简单。
Reconstructing Humans and Objects in Interaction using Large Reconstruction Models
- 利用大重建模型的几何结构作为骨架，简化3D交互恢复。
- 在多个数据集上优于现有方法，尤其在遮挡和形变场景下表现好。
- 适合做虚拟现实、机器人交互等需要精准3D理解的任务。
3D人体-物体交互（3D HOI）是3D计算机视觉的基础问题，广泛应用于AR/VR、机器人和具身AI。然而，由于深度歧义、遮挡和物体形状变化，其3D重建仍具挑战性。现有方法主要依赖2D投影和接触约束，将参数化人体模型和物体模板拟合到图像。本文提出MILO框架，利用大重建模型（LRMs）的视觉能力，从单张图像中恢复详细的3D人物交互。关键观察是：LRMs提供强大的几何骨架，能保持人与物体间的相对位置和距离线索，显著简化重建流程。我们将该模型网格分割为人体与物体部分，对人体部分拟合参数化人体模型，并可选地对物体部分对齐物体模板（若有）。MILO在多个基准和交互场景中实现强重建精度，超越现有基线。代码已开源：https://ac5113.github.io/MILO。
原文摘要 · Abstract (English)
Estimation of Human-Object Interactions in 3D (3D HOI) is a fundamental problem in 3D computer vision with applications in AR/VR, robotics, and embodied AI. However, reconstructing these interactions in 3D remains challenging due to depth ambiguities, occlusions, and object shape variability. Existing approaches are primarily concerned with reprojection and contact constraints, fitting parametric human models and object templates to 2D images. In this paper, we explore a different avenue. We present MILO, a framework that leverages the visual capabilities of Large Reconstruction Models (LRMs) to recover detailed 3D human-object interactions from a single image. Our key observation is that LRMs provide a powerful geometric scaffold that preserves relative human-object arrangement and proximity cues. This significantly simplifies the reconstruction procedure, reframing the problem as interpreting the LRM mesh: we segment it into human and object components, fit a parametric body model to the human part, and optionally align an object template to the object part (if such a template is available). MILO achieves strong reconstruction accuracy and outperforms existing baselines across multiple benchmarks and interaction scenarios. Our code is available at https://ac5113.github.io/MILO.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。