A Benchmark for Temporal Color Constancy
Yanlin Qian, Jani Käpylä, Joni-Kristian Kämäräinen, Samu Koskinen, Jiri Matas (ArXiv, 2020)
The benchmark dataset comprises 600 real-world sequences recorded with a high-resolution mobile phone camera and a fixed train-test split which ensures consistent evaluation. The data is divided in a training-test split comprising 400 items for training and 200 for testing. Both training and test data are stored in multiple compressed archives. Further instructions on data preprocessing are available on the official repository of the paper on GitHub.