基于扩散的循环一致性学习实现超越配对数据稀缺的高保真虚拟试衣
High-Fidelity Virtual Try-On beyond Paired Data Scarcity via Diffusion-based Cycle-Consistent Learning
提出基于扩散模型循环一致性学习的虚拟试衣框架，利用无配对数据训练统一生成模型，其多准则过滤和感知正则化策略可迁移到其他需要无配对数据的图像编辑任务。
原文依据 we propose Cycle-Consistent Virtual Try-On (CCVTON), a diffusion-based approach that enables effective training using massive in-the-wild person images.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论基于扩散的虚拟试衣方法依赖大量高质量的服装-人物配对数据，而由于数据收集和预处理成本高昂，这类数据在实践中十分稀缺，限制了其在真实场景中的性能。为克服这一瓶颈，我们提出了循环一致性虚拟试衣（CCVTON），一种基于扩散的方法，能够利用大量野外人物图像进行有效训练。具体而言，CCVTON 引入了一种循环一致性学习（CCL）策略，仅使用一个统一的生成模型将服装从人物图像中解耦（脱衣分支）并将其转移到同一个人身上（试衣分支），形成重建循环。为此，我们首先在开源配对数据上预热一个统一扩散 Transformer（UDiT），以获得基本的试衣和脱衣能力。当将 UDiT 适应到野外人物图像时，我们采用多准则过滤操作来选择由预训练 UDiT 从人物图像中解耦出的高质量服装。这些过滤后的服装不作为 CCL 的输入，而是作为感知正则化损失的软约束，防止脱衣分支退化为简单的复制。此外，我们提出了具有两阶段细化过程的服装感知掩码生成，以抑制服装泄漏同时保持人物一致性。大量实验表明，CCVTON 达到了最优结果。
英文原文摘要 1434 字符
Diffusion-based virtual try-on methods rely on vast high-quality garment-person pairs, which are scarce in practice due to the high cost of data collection and preprocessing, limiting their performance in real-world scenarios.To overcome this bottleneck, we propose Cycle-Consistent Virtual Try-On (CCVTON), a diffusion-based approach that enables effective training using massive in-the-wild person images. Specifically, CCVTON introduces a Cycle-Consistent Learning (CCL) strategy that just employs a single unified generative model to disentangle a garment from a person image (try-off branch) and transfer it to the same individual (try-on branch), forming a reconstruction cycle. To this end, we first warm up a Unified Diffusion Transformer (UDiT) on open-source paired data to acquire basic try-on and try-off capabilities. When adapting UDiT to in-the-wild person images, we employ a Multi-Criteria Filtering Operation to select high-quality garments disentangled from person images by the pretrained UDiT. These filtered garments are not used as inputs for CCL, but serve as soft constraints for a perceptual regularization loss, preventing the try-off branch from collapsing to trivial copying. In addition, we propose garment-aware mask generation with a two-stage refinement process to suppress garment leakage while maintaining person consistency. Extensive experiments show that CCVTON achieves state-of-the-art results.