You need to agree to share your contact information to access this dataset
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
This dataset is available exclusively for non-commercial academic research. You must confirm that you already have access to the original DressCode dataset, and use an institutional email to request access. By requesting access, you agree NOT to use this dataset or its content for any commercial purposes, and NOT to redistribute any part of it publicly or privately.
Log in or Sign Up to review the conditions and access this dataset content.
📦 DressCode-MR
DressCode-MR is a high-quality multi-reference virtual try-on dataset built upon the original DressCode dataset. It is designed for research on multi-reference garment transfer, diffusion-based try-on, and layout-to-appearance modeling.
🔍 Description
DressCode-MR constructs a rich set of try-on samples by:
- Selecting valid human-garment pairs from the original DressCode dataset;
- Generating multi-reference garment & accessory image sets for each try-on case;
- Leveraging multiple expert models based on CatVTON-FLUX to perform inverse try-on (TryOff), producing high-fidelity reference images;
- Manually filtering for high-quality and diverse garment/accessory-person combinations.
📊 Dataset Statistics
| Split | Samples |
|---|---|
| Train | 25,779 |
| Test | 2,400 |
| Total | 28,179 |
📄 Citation
If you use this dataset, please cite the following work:
- Downloads last month
- 5