multimodal active speaker detection and virtual cinematography for video conferencing

• DOI: 10.48550/arxiv.2002.03977
• Publication Date: 2020-05-01
• Authors: Cha Zhang; Ramin Mehran; Oliver A. Whyte; Adarsh Kowdle; Ross Cutler; Adam G. Kirk; Samuel Alan Johnson
• Keywords: FOS: Computer and information sciences; Computer Science - Machine Learning; Machine Learning (stat.ML); 02 engineering and technology; Machine Learning (cs.LG); Multimedia (cs.MM); Statistics - Machine Learning; Audio and Speech Processing (eess.AS); FOS: Electrical engineering, electronic engineering, information engineering; 0202 electrical engineering, electronic engineering, information engineering; Computer Science - Multimedia; Electrical Engineering and Systems Science - Audio and Speech Processing

Active speaker detection (ASD) and virtual cinematography (VC) can significantly improve the remote user experience of a video conference by automatically panning, tilting and zooming of a video conferencing camera: users subjectively rate an expert video cinematographer's video significantly higher than unedited video. We describe a new automated ASD and VC that performs within 0.3 MOS of an expert cinematographer based on subjective ratings with a 1-5 scale. This system uses a 4K wide-FOV camera, a depth camera, and a microphone array; it extracts features from each modality and trains an ASD using an AdaBoost machine learning system that is very efficient and runs in real-time. A VC is similarly trained using machine learning to optimize the subjective quality of the overall experience. To avoid distracting the room participants and reduce switching latency the system has no moving parts -- the VC works by cropping and zooming the 4K wide-FOV video stream. The system was tuned and evaluated using extensive crowdsourcing techniques and evaluated on a dataset with N=100 meetings, each 2-5 minutes in length.