AI Virtual Try-On · Computer Vision R&D
Garment transfer that preserves pose and fabric detail.
A diffusion-based virtual try-on pipeline that transfers a garment onto a person photo while preserving pose and fabric detail, built for e-commerce.
This work was built for a private engagement, so the code and full pipeline aren't public. What follows is the approach at a level I can share, with a recorded result.
- Type
- Computer vision R&D
- Domain
- E-commerce visualization
- Input
- Person photo + garment
- Source
- Private
A recorded run.
The pipeline, in stages.
Human parsing
Segment the person into semantically consistent regions so garment placement respects body parts and clothing boundaries.
Pose extraction
Estimate keypoints so the transferred garment follows the person's stance instead of the reference garment's pose.
Garment transfer
Warp the target garment onto the person with pose-conditioned alignment.
Diffusion refinement
A refinement pass restores fabric texture, folds, and lighting consistency that geometric warping flattens out.
Hard parts.
Most of the difficulty was between the models: keeping garment geometry stable across poses, preserving fine patterns that parsing blurs, and keeping the person's identity through the diffusion pass. Results were judged on typical e-commerce cases such as sleeves, prints, and layered clothing.