• Description:

CLIC is a dataset for the Challenge on Learned Image Compression 2020 lossy image compression track. These images contain a mix of the professional and mobile datasets used to train and benchmark rate-distortion performance. The dataset contains both RGB and grayscale images. This may require special handling if a grayscale image is processed as a 1 channel Tensor and a 3 channel Tensor is expected.

This dataset does NOT contain the data from the P-Frame challenge (YUV image frames).

Split Examples
'test' 428
'train' 1,633
'validation' 102
  • Feature structure:
    'image': Image(shape=(None, None, 3), dtype=uint8),
  • Feature documentation:
Feature Class Shape Dtype Description
image Image (None, None, 3) uint8


  • Citation:
  title = {Workshop and Challenge on Learned Image Compression (CLIC2020)},
  author = {George Toderici, Wenzhe Shi, Radu Timofte, Lucas Theis,
            Johannes Balle, Eirikur Agustsson, Nick Johnston, Fabian Mentzer},
  url = {},