OASIS: Occlusion-aware Single-image Hand Avatar Reconstruction via 3D Gaussian Splatting
2026-07-31 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed OASIS, a new way to create 3D hand models from just one picture, which is usually very hard because hands can block themselves and have many moving parts. They improve on previous methods by using 3D Gaussian Splatting that better handles fine details and complex hand shapes. Their method aligns image details with 3D hand geometry and uses a smart attention system to manage parts that are blocked or hidden. They also introduce a way to reflect how the hand surface stretches and move accordingly. Tests show their approach works well and is fast, and it can be used for making hand avatars from text or editing textures.
Single-image 3D reconstruction3D Gaussian SplattingSelf-occlusionHand avatarVisual evidence tokensVisibility-conditioned attentionFeature-on-Mesh representationNon-rigid deformationOne-shot adaptationText-to-avatar generation
Authors
Zhisheng Han, Shiyao Wu, Jiayan Qiu, Yakun Ju, Lu Liu, Le Zhang, Pengfei Feng, Huiyu Zhou, Zheheng Jiang
Abstract
Single-image 3D hand avatar reconstruction is fundamentally ill-posed and particularly challenging due to limited visual evidence under severe self-occlusion and the complex pose-dependent deformation of highly articulated hands. Existing methods predominantly rely on implicit NeRF-style representations, whose volumetric fitting is computationally expensive and often struggles to preserve fine-grained hand details. In this work, we present OASIS, a tailored 3D Gaussian Splatting framework for single-image hand avatar reconstruction. To faithfully encode sparse image-specific appearance cues in single-view reconstruction, we construct geometry-aligned visual evidence tokens by explicitly aligning input image observations with 3D hand geometry and context-adaptively tokenizing the resulting visual evidence. Since severe self-occlusion makes the reliability of image evidence inherently visibility-dependent, we introduce a visibility-conditioned point-image attention to reliably transfer visual evidence to geometric tokens, yielding occlusion-aware Gaussian features for faithful and robust reconstruction. To further capture non-rigid deformation of articulated hands, we introduce a Feature-on-Mesh representation to enable Gaussian deformation to be guided by local surface stretching. Under this framework, we adopt a one-shot adaptation scheme that learns a shared hand prior from multi-identity training data and then fits it to a target image for target-specific reconstruction. Extensive experiments show that OASIS outperforms existing baselines in both visual fidelity and efficiency across challenging poses and in-the-wild scenarios, and further demonstrates strong versatility in downstream applications such as text-to-avatar generation and texture editing.