This repo contains the dataset anon-annotationsv1, which is used for training anon, and benchmarks for evaluating anon. The data distribution of anon-annotationsv1 is as follows:

Please refer our github page for the usage.