This site is a platform for all information about automated action recognition and classification. It provides an overview of current benchmark datasets, results, papers, code and many more informations related to action recognition. We try to build a place to go for everyone ... students, researchers, authors, reviewers ... looking for latest results, paper overviews and state-of-the-art implementations related to action recognition from video data.

If you're an author of a new action recognition paper we didn't list yet, please use the 'Add new Paper' button to add your paper and results to the overviews. As we're still in the beginning, please let us know whatever you like or dislike on this site, what we can improve or what else might help you in your work.

Please use all information provided on this site carefully and responsibly!


If you're planing to use information provided on this site, please keep in mind that all numbers and papers are added by authors without double checking. We of course try to keep results as accurate as possible, and whenever we got notice of an error it will be fixed, but this does not release you from the obligation of reading the papers and double checking the numbers listed here before using them.

If you want a monthly update on newly added papers, please join our mailing list: