2016-08-08 The data and labels of the attribute prediction benchmark are released without encryption (password). If Dropbox is not accessible, please download the dataset using Google Drive or Baidu Drive


Category and Attribute Prediction Benchmark evaluates the performance of clothing category and attribute prediction. This is a large subset of DeepFashion, containing massive descriptive clothing categories and attributes in the wild. It contains

  • 289,222 number of clothes images;

  • 50 number of clothing categories, and 1,000 number of clothing attributes;

  • Each image is annotated by bounding box and clothing type.


  • Clothes Images

  • High-Res Images

  • Bounding Box Annotations

  • Category Annotations

  • Landmark Annotations

  • Attribute Annotations

  • Fine-Grained Attribute Annotations

  • Train/Val/Test Partitions


 author = {Liu, Ziwei and Luo, Ping and Qiu, Shi and Wang, Xiaogang and Tang, Xiaoou},
 title = {DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations},
 booktitle = {Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 month = {June},
 year = {2016}


Please contact Ziwei Liu for questions about the dataset.