ICTCAS-UCAS-TAL Submission to the AVA-ActiveSpeaker Task at ActivityNet Challenge 2021
This report presents a brief description of our method for the AVA Active Speaker Detection (ASD) task at ActivityNet Challenge 2021. Our solution, the Extended Unified Context Network (Extended UniCon) is based on a novel Unified Context Network (UniCon) designed for robust ASD, which combines multiple types of contextual information to optimize all candidates jointly. We propose a few changes to the original UniCon in terms of audio features, temporal modeling architecture, and loss function design. Together, our best model ensemble sets a new state-of-the-art at 93.4% mAP on the AVA-ActiveSpeaker test set without any form of pretraining, and currently ranks first on the ActivityNet challenge leaderboard.
PDFDatasets
Results from the Paper
Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
---|---|---|---|---|---|---|
Audio-Visual Active Speaker Detection | AVA-ActiveSpeaker | Extended UniCon | validation mean average precision | 93.6% | # 6 |