Disentangled Clothed Avatar Generation from Text Descriptions
This addresses the challenge of entangled avatar representations for downstream applications like animation and editing in computer graphics and virtual reality, offering a novel method but with incremental improvements over existing SMPL-based approaches.
The paper tackles the problem of generating 3D avatars from text descriptions by proposing a disentangled representation that separates the human body and clothes, enabling high-quality animation and editing. The result is improved texture and geometry quality, better semantic alignment with text, and enhanced visual performance in tasks like character animation and virtual try-on.
In this paper, we introduce a novel text-to-avatar generation method that separately generates the human body and the clothes and allows high-quality animation on the generated avatar. While recent advancements in text-to-avatar generation have yielded diverse human avatars from text prompts, these methods typically combine all elements-clothes, hair, and body-into a single 3D representation. Such an entangled approach poses challenges for downstream tasks like editing or animation. To overcome these limitations, we propose a novel disentangled 3D avatar representation named Sequentially Offset-SMPL (SO-SMPL), building upon the SMPL model. SO-SMPL represents the human body and clothes with two separate meshes but associates them with offsets to ensure the physical alignment between the body and the clothes. Then, we design a Score Distillation Sampling (SDS)-based distillation framework to generate the proposed SO-SMPL representation from text prompts. Our approach not only achieves higher texture and geometry quality and better semantic alignment with text prompts, but also significantly improves the visual quality of character animation, virtual try-on, and avatar editing. Project page: https://shanemankiw.github.io/SO-SMPL/.