minar09/awesome-virtual-try-on

What it solves

This project provides a comprehensive, organized index of research and tools focused on Virtual Try-on (VTON). It helps researchers and developers navigate the rapidly evolving landscape of AI-powered clothing synthesis, covering everything from basic image-based swapping to complex 3D and video-based try-on technologies.

How it works

The repository acts as a curated directory of academic papers, open-source codebases, datasets, and project demos. It categorizes these resources by their underlying technology (such as Diffusion Models or ControlNet) and specific application areas (such as multi-pose guidance, in-the-wild scenarios, or video synthesis), allowing users to find relevant state-of-the-art methods for specific virtual dressing tasks.

Who it’s for

This list is designed for researchers, computer vision engineers, and developers interested in fashion-related AI, generative modeling, and human image synthesis who want a centralized source to track advancements in the VTON field.

Highlights

• Extensive collection of research papers from top-tier conferences like CVPR, ICCV, ECCV, and NeurIPS.
• Categorization by technical approach, including prompt-based, 2D image-based, 3D, and video-based try-on.
• Inclusion of practical resources like datasets, safety and moderation tools, and live project demos.
• Regular updates featuring the latest developments in generative AI and diffusion-based clothing synthesis.