Dataset · Layered-Labs
Facial photos with acne and severity labels
Layered-Labs/ACNE04
A corpus of 1,457 images with Hayashi scale acne severity assessment and nearly 19k lesion boxes.
What's inside
A reassembled version of the original ACNE04 dataset containing 1,457 facial photographs with a clean separation of images and labels. Each image has a verified Hayashi 4-level severity assessment, plus 18,983 bounding boxes for individual rash elements. All images are embedded directly in the dataset for fast loading, with English description.
What you can do with it
- Training object detectors (YOLO, Faster R-CNN) for skin lesion detection and counting
- Training classifiers for automated acne severity assessment
- Creating AI assistant tools for telemedicine and dermatological screening
- Tasks
- Image classification, Object detection
- Languages
- English
- Size
- 1K–10K rows
- Downloads this month
- 1.5K
- Likes
- 4
How to load it
from datasets import load_dataset
ds = load_dataset("Layered-Labs/ACNE04")