# Human Interaction-Aware 3D Reconstruction from a Single Image Source: [Human Interaction-Aware 3D Reconstruction from a Single Image]() Collection index: [llms.txt]() Data updated: 2026-04-09T05:00:04.534Z Authors: Gwanghyun Kim, Junghun James Kim, Suh Yoon Jeon, Jason Park, Se Young Chun Published: 2026-04-07T05:09:41.000Z ## Abstract Reconstructing textured 3D human models from a single image is fundamental for AR/VR and digital human applications. However, existing methods mostly focus on single individuals and thus fail in multi-human scenes, where naive composition of individual reconstructions often leads to artifacts such as unrealistic overlaps, missing geometry in occluded regions, and distorted interactions. These limitations highlight the need for approaches that incorporate group-level context and interaction priors. We introduce a holistic method that explicitly models both group- and instance-level information. To mitigate perspective-induced geometric distortions, we first transform the input into a canonical orthographic space. Our primary component, Human Group-Instance Multi-View Diffusion (HUG-MVD), then generates complete multi-view normals and images by jointly modeling individuals and group context to resolve occlusions and proximity. Subsequently, the Human Group-Instance Geometric Reconstruction (HUG-GR) module optimizes the geometry by leveraging explicit, physics-based interaction priors to enforce physical plausibility and accurately model inter-human contact. Finally, the multi-view images are fused into a high-fidelity texture. Together, these components form our complete framework, HUG3D. Extensive experiments show that HUG3D significantly outperforms both single-human and existing multi-human methods, producing physically plausible, high-fidelity 3D reconstructions of interacting people from a single image. Project page: https://jongheean11.github.io/HUG3D\_project ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Multi-human 3D reconstruction on MultiHuman](): Multi-human 3D reconstruction; MultiHuman; CD: 3.631. - [3D Human Reconstruction (Occluded Regions) on MultiHuman Closely](): 3D Human Reconstruction (Occluded Regions); MultiHuman Closely; L2 Norm Error: 0.153. - [3D Human Reconstruction (Occluded Regions) on MultiHuman Natural](): 3D Human Reconstruction (Occluded Regions); MultiHuman Natural; Norm L2: 0.138. - [3D Multi-Human Reconstruction on MultiHuman Closely interaction](): 3D multi-human reconstruction; MultiHuman Closely Interaction; CD: 4.315. - [3D Multi-Human Reconstruction on MultiHuman Natural interaction](): 3D multi-human reconstruction; MultiHuman Natural Interaction; CD: 3.34. - [Multi-human 3D reconstruction on MultiHuman occluded region](): Multi-human 3D reconstruction; MultiHuman occluded region; Norm L2: 0.14. - [3D Human Reconstruction on MultiHuman Closely Interaction](): 3D human reconstruction; MultiHuman Closely Interaction; PSNR: 16.454. - [3D Human Reconstruction on MultiHuman Natural](): 3D human reconstruction; MultiHuman Natural Interaction; PSNR: 16.741. - [Multi-human 3D Reconstruction on MultiHuman (PSNR, SSIM, LPIPS)](): Multi-human 3D reconstruction; MultiHuman; PSNR: 16.456. - [SMPL-X fitting on MultiHuman](): SMPL-X fitting; MultiHuman; MPJPE: 13.139. - [Runtime and VRAM for Multi-human 3D Reconstruction on Multi-human 3D Reconstruction dataset](): Multi-human 3D reconstruction; Multi-human 3D Reconstruction dataset; Elapsed Time (s): 216.32. ## Original sources and implementation [Original paper]() [PDF]()