DeepFashion2: A Versatile Benchmark for Detection, Pose Estimation, Segmentation and Re-Identification of Clothing Images

Yuying Ge, Ruimao Zhang, Xiaogang Wang, Xiaoou Tang, Ping Luo; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 5337-5345

Abstract


Understanding fashion images has been advanced by benchmarks with rich annotations such as DeepFashion, whose labels include clothing categories, landmarks, and consumer-commercial image pairs. However, DeepFashion has nonnegligible issues such as single clothing-item per image, sparse landmarks (4 8 only), and no per-pixel masks, making it had significant gap from real-world scenarios. We fill in the gap by presenting DeepFashion2 to address these issues. It is a versatile benchmark of four tasks including clothes detection, pose estimation, segmentation, and retrieval. It has 801K clothing items where each item has rich annotations such as style, scale, view- point, occlusion, bounding box, dense landmarks (e.g. 39 for 'long sleeve outwear' and 15 for 'vest'), and masks. There are also 873K Commercial-Consumer clothes pairs. The annotations of DeepFashion2 are much larger than its counterparts such as 8x of FashionAI Global Challenge. A strong baseline is proposed, called Match R- CNN, which builds upon Mask R-CNN to solve the above four tasks in an end-to-end manner. Extensive evaluations are conducted with different criterions in Deep- Fashion2. DeepFashion2 Dataset will be released at : https://github.com/switchablenorms/DeepFashion2

Related Material


[pdf]
[bibtex]
@InProceedings{Ge_2019_CVPR,
author = {Ge, Yuying and Zhang, Ruimao and Wang, Xiaogang and Tang, Xiaoou and Luo, Ping},
title = {DeepFashion2: A Versatile Benchmark for Detection, Pose Estimation, Segmentation and Re-Identification of Clothing Images},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}