Dataset · mattmdjaga
Human and clothing segmentation (ATR copy)
mattmdjaga/human_parsing_dataset
17,706 image and mask pairs for pixel-level segmentation of clothing, accessories, and human body parts.
What's inside
It contains 17,706 images with corresponding segmentation masks. This is a copy of the well-known Deep Human Parsing ATR dataset. The markup covers clothing elements and body parts, including hats, hair, glasses, upper and lower clothing, shoes, face, and legs.
What you can do with it
- Training clothing semantic segmentation models for virtual fitting rooms
- Detecting and extracting body parts and accessories in photos
- Preparing masks for image editing and inpainting tasks
- Tasks
- Image segmentation
- Size
- 10K–100K rows
- Downloads this month
- 1K
- Likes
- 50
Models trained on it
- mattmdjaga/segformer_b2_clothesImage segmentation 326.3K
- sayeed99/segformer_b3_clothesImage segmentation 26.1K
- mattmdjaga/segformer_b0_clothesImage segmentation 765
- isjackwild/segformer-b0-finetuned-segments-skin-hair-clothingImage segmentation 247
- gj12138/segformer_b2_clothesImage segmentation 15
- laohcui/segformer_b2_clothesImage segmentation 11
- VanNguyen1214/get_face_and_hairImage segmentation 8
- devilking567/segformer_b2_clothesImage segmentation 8
How to load it
from datasets import load_dataset
ds = load_dataset("mattmdjaga/human_parsing_dataset")