发表机构
Key Laboratory of Media Audio and Video (Communication University of China); School of Intelligence Science and Technology, Nanjing University（中国传媒大学媒体音频与视频重点实验室; 南京大学智能科学与技术学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出DiGS-Avatar，将单图像可动画三维人体重建转化为UV空间扩散的潜变量补全任务，通过师生框架优化后解码为三维高斯基元，实现了高效高质量的重建与零样本泛化。
AI 中文摘要
单图像三维人体重建常存在纹理过平滑与几何不一致问题。扩散模型虽能提升生成质量，但依赖多视图合成先验进行三维重建的计算成本高且易出现视图不一致。本文提出DiGS-Avatar，将该任务重新定义为高效的、基于扩散的UV潜变量补全任务，从设计上保证三维一致性。为捕捉准确的空间结构，引入师生框架：多视图教师提供几何对齐的伪真值潜变量，用于监督单视图扩散学生。将该推断出的潜变量作为鲁棒结构骨架，本文方法注入高级语义特征以准确恢复精细纹理细节，同时不破坏空间完整性。最终将优化后的表示解码为三维高斯基元。大量实验表明，DiGS-Avatar达到了SOTA或极具竞争力的视觉保真度与零样本泛化能力，且仅需0.71秒即可重建出完全可动画的三维化身。代码可在指定URL获取。
英文摘要
Single-image 3D human reconstruction often suffers from over-smoothed textures and geometric inconsistencies. While diffusion models improve generative quality, their reliance on multi-view synthesis prior to 3D reconstruction is computationally expensive and prone to view inconsistency. We propose DiGS-Avatar, which reformulates this task as an efficient, diffusion-based UV-latent completion task, ensuring 3D consistency by design. To capture accurate spatial structure, we introduce a teacher-student framework where a multi-view teacher provides geometrically aligned pseudo-ground-truth latents to supervise a single-view diffusion student. Treating this inferred latent as a robust structural skeleton, our method injects high-level semantic features to accurately recover fine textural details without disrupting spatial integrity. The refined representation is then decoded into 3D Gaussian primitives. Extensive experiments demonstrate that DiGS-Avatar achieves state-of-the-art or highly competitive visual fidelity and zero-shot generalization, while reconstructing a fully animatable 3D avatar in just 0.71 seconds. Code is available at https://github.com/KLMAV-CUC/DiGS-Avatar.
CommentsECCV 2026
