• Description:

Caltech-101 consists of pictures of objects belonging to 101 classes, plus one background clutter class. Each image is labelled with a single object. Each class contains roughly 40 to 800 images, totalling around 9k images. Images are of variable sizes, with typical edge lengths of 200-300 pixels. This version contains image-level labels only. The original dataset also contains bounding boxes.

Split Examples
'test' 6,084
'train' 3,060
  • Feature structure:
    'image': Image(shape=(None, None, 3), dtype=uint8),
    'image/file_name': Text(shape=(), dtype=string),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=102),
  • Feature documentation:
Feature Class Shape Dtype Description
image Image (None, None, 3) uint8
image/file_name Text string
label ClassLabel int64


  • Citation:
  title={Learning Generative Visual Models from Few Training Examples: An Incremental Bayesian Approach Tested on 101 Object Categories},
  author={Li Fei-Fei and Rob Fergus and Pietro Perona},
  journal={Computer Vision and Pattern Recognition Workshop},