About join the activityNet community

[陳炫均]

Hi, I am planing to research the task of audiovisual active speaker detection using ActivityNet. But there are only train and val set publish now. Can I send the evaluation result for you to help me get the result to compare with other existing SOTA model ? By the way, unfortunately, I miss the challenge this year.