Automatic camera control for capturing collaborative meetings

The growing size of organizations is making it increasingly expensive to attend\nmeetings and difficult to retain what happened in those meetings. Meeting video\ncapture systems exist to support video conferencing for remote participation or\narchiving for later review, but they have been regarded ineffective. The reason\nis twofold. Firstly, the conventional way of capturing video using a single static\ncamera fails to capture focus and context. Secondly, a single static view is often\nmonotonous, making the video onerous to review. To address these issues, often\nhuman camera operators are employed to capture effective videos with changing\nviews, but this approach is expensive.\nIn this thesis, we argue that camera views can be changed automatically to\nproduce meeting videos effectively and inexpensively. We automate the camera view control by automatically determining the visual focus of attention as a function\nof time and moving the camera to capture it. In order to determine visual\nfocus of attention for different meetings, we conducted experiments and interviewed\ntelevision production professionals who capture meeting videos. Furthermore,\ntelevision production principles were used to appropriately frame shots\nand switch between shots.\nThe result of the evaluation of the automatic camera control system indicated\nits significant benefits over conventional static camera view. By applying television\nproduction principles various issues related to shot stability and screen\nmotion were resolved. The performance of the automatic camera control based\non television production principles also approached the performance of trained\nhuman camera crew. To further reduce the cost of the automation, we also explored\nthe application of computer vision and audio tracking.\nResults of our explorations provide empirical evidence in support of the utility\nof camera control encouraging future research in this area. Successful application\nof television production principles to automatically control cameras suggest\nvarious ways to handle issues involved in the automation process.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.