SceneNet FET project organized the event which revolved around the Gestalt concept: taking multiple audio-visual sources and creating a better, richer experience that can be shared among users. Along with the organizers, eight European projects actively participated in the Audio-Visual Gestalt Workshop:
In order to achieve goal the Audio-Visual Gestalt experience, there is a need to provide technology to gather information from multiple sources, analyze and synchronize this information and then create an unified informative output. The development and implementation of state of the art algorithms in computer vision and audio analysis is needed, requiring immense computing power. Each project handles this multilayered concept from a different viewpoint and aims at different achievements. Consequently, during the workshop, several aspects of this topic were discussed:
• data acquisition
• data synchronization
• 3D/4D visual data reconstruction
• compute related issues
• visualization aspects
• ethical, legal and social questions that arise from this concept
The introductions to all participating projects by their representatives was followed by lively Round Table Discussions with the topics “Acquisition and data flow”, “Collaborative processing” as well as “User experience and visualization of audio-visual information”.
The invited talk by Prof. Dr. Franziska Boehm, Munster, Germany, a new member of the SceneNet Advisory Board, about “Legal aspects of crowd sourcing of audio-visual information” led to very detailed discussions about ethical, social and legal aspects in the joint research topics.
The fruitful discussions, the gathering of so many different people with different thoughts and research backgrounds and in addition the overall conclusion to aim for further future collaboration had a positive impact on the SceneNet project and people in the SceneNet consortium.