DeepFashion2 Clothing Dataset

* This dataset supports online use.Click here to jump.

DeepFashion2 is a general benchmark that includes four tasks: clothing detection, pose estimation, segmentation, and retrieval. The dataset includes 801,000 clothing items, each with rich annotations such as style, scale, viewpoint, occlusion, bounding box, dense landmark, and mask. In addition, it also contains 873,000 commercial-consumer clothing pairs.