Active Speaker Detection as a Multi-Objective Optimization with Uncertainty-based Multimodal Fusion
This work addresses the problem of unreliable modalities in audiovisual fusion for active speaker detection, which is incremental as it builds on existing multimodal approaches.
The paper tackles active speaker detection by framing it as a multi-objective optimization problem with a novel uncertainty-based multimodal fusion scheme, resulting in improved mAP and AUC scores and surpassing state-of-the-art on the AVA-ActiveSpeaker dataset.
It is now well established from a variety of studies that there is a significant benefit from combining video and audio data in detecting active speakers. However, either of the modalities can potentially mislead audiovisual fusion by inducing unreliable or deceptive information. This paper outlines active speaker detection as a multi-objective learning problem to leverage best of each modalities using a novel self-attention, uncertainty-based multimodal fusion scheme. Results obtained show that the proposed multi-objective learning architecture outperforms traditional approaches in improving both mAP and AUC scores. We further demonstrate that our fusion strategy surpasses, in active speaker detection, other modality fusion methods reported in various disciplines. We finally show that the proposed method significantly improves the state-of-the-art on the AVA-ActiveSpeaker dataset.