A multi-modal dataset containing images with segmentation masks and bounding boxes for 37 specific cat and dog breeds.
Dataset Details
Data Fields
img: RGB image
msk: Grayscale image (0: foreground, 1: background, 2: ambiguous)
bbox: Sequence of integers representing bounding boxes in [x_min, y_min, width, height] format
class: Binary class label (0=cat, 1=dog)
category: Fine-grained breed classification from 37 classes:
Cats (17 breeds):… See the full description on the dataset page: https://huggingface.co/datasets/cvdl/oxford-pets.
