HS Düsseldorf >> 
Fachbereiche >> 
Institute >> 
Organe und Gremien >> 
Verwaltung >> 
Hochschulbibliothek >> 
Campus IT >> 
Suche & Hilfe >> 

FB 1 Architektur >> 
FB 2 Design >> 
FB 3 Elektro- & Informationstechnik >> 
FB 4 Maschinenbau >> 
FB 5 Medien >> 
FB 6 Sozial/Kulturwiss. >> 
FB 7 Wirtschaft >> 

Aktuelles >> 
Studium >> 
Personen >> 
Showroom >> 
Studienbüro >> 
Dekanat >> 
Internationales >> 
Forschung & Entwicklung  >> 

Projekte >> 
Labore, Pools & Studios  >> 


Hochschule Düsseldorf
University of Applied Sciences


Fachbereich Medien
Faculty of Media

Labor für Virtuelles Studio / Virtuelle Realität

Publikationen >>
Presentationen >>
Forschung >>
Technologie >>
Kooperationen >>
Galerie >>
Abschlussarbeiten >>
Events >>
Lehre >>
Anfahrt >>
Labor Suche >>
Aktuelles >>

Automatic Camera and Switcher Control for Live Broadcast using Scene Analysis and Expert System (2015)


Snapshot: Campus Derendorf Snapshot: Library and Campus IT Snapshot: Canteen and Administration Snapshot: Faculty of Social Sciences & Cultural Studies; Faculty of Business Studies
Snapshot: Faculty of Media Snapshot: Faculty of Design; Faculty of Architecture Snapshot: Faculty of Electrical & Information Technology Snapshot: Faculty of Mechanical & Process Engineering
Snapshot: Virtual Studio Snapshot: Andrea Gerhard Snapshot: Apex(a,b) Snapshot: Professor Jens Herder
Snapshot: Ext(a,b) Snapshot: miniature model explaining the scene Snapshot: Pan(a) Snapshot: Andrea Gerhard points to Jibarm


Live video broadcasting requires a multitude of professional expertises to enable real-time multi-camera productions. Unlike manually operated cameras, robotic systems allow the automation of common and repeated tracking shots. However, predefined camera shots do not allow quick adjustments when required due to unpredictable events. In this paper we introduce a modular automated camera control and video switch system, based on fundamental cinematographic rules and principles as they are used in feature film and studio productions. By using the positional information of the actors provided by a markerless tracking system and the sound level of lavalier microphones, the current situation in the scene is analyzed by an expert system, which determines appropriate camera angles and decides when to switch from one camera to another. Based on this evaluation and the physical constraints of the robotic camera, low-level modules operate the robotic camera to establish the appropriate shot, and control the video switcher, which performs the required transitions. A test production was done to observe the developed prototype in a live broadcast scenario and served as a video-demonstration for an evaluation based on questionnaires and interviews with experts to show possible challenges and suggestions for further developments.




The media player is loading...




Description Time mp4
Original09:52 Min 454 MB >> 



Evaluation Examples


The media player is loading...


Production team 2015


Following members took part in the production: Ulrich Braas, Jose Burga, Jeff Daemen, Maximilian Friedrich, Terence Geldner, Andrea Gerhard, Jens Herder, Julian Jochheim, Alexander Jussen, Mario Kascholke, Daniel Kirchhof, Cornelius Koch, Okan Köse, Philipp Ladwig, Dennis Meyer, Antje Müller, Marcel Tiator, Roman Wiche, Kai Wilgen


Application of a Moderator Tracking for Camera Control in a Virtual (TV) Studio (2014)


Snapshot: Flug durch das VSVR Logo Snapshot: Arena Snapshot: Schiedsrichter Snapshot: Program 1 - Cyborg
Snapshot: Program 2 - Bot Mixed Reality Labor: Steuerung des Avatars Snapshot: Avatar wirft Disk mit Schweif Snapshot: Cyborg fängt Disk


Markerless actor tracking for Virtual TV Studios enables sophisticated real-time robotic camera control. We demonstrate a semi-automatic prototype, which uses cinematographic rules in combination with the tracked positions of the actors to determine appropriate camera angles. The operator of a manually operated camera is able to switch the playout between automatic and manual camera by interacting with a mobile device. This trigger is also used to assign the determined parameters to the robotic camera. Two different shot types, the “over-the-shoulder” and “master” shot, were implemented. We discuss advantages and problems of such a system for standard formats by evaluating a test production.




The media player is loading...




Description Time mp4
Original06:43 Min 200 MB >> 


Production team 2014


Following members took part in the production: Frano Barnjak, Jeff Daemen, Moritz Heilemann, Jens Herder, Philipp Ladwig, Niklas Petau, Mitja Säger, Margareta Urbaniak



Semi-Automatic Camera and Switcher Control for Live Broadcast
Jeff Daemen, Jens Herder, Cornelius Koch, Philipp Ladwig, Roman Wiche, and Kai Wilgen, International Conference on Interactive Experiences for Television and Online Video (TVX'2016), Chicago, IL, USA, ACM, DOI=10.1145/2932206.2933559, June 22-24, 2016. [wip poster]
Markerless Actor Tracking for Virtual (TV) Studio Applications
Production at the Virtual Studio of FH Düsseldorf, 2013.
Markerless Actor Tracking for Virtual (TV) Studio Applications
Jeff Daemen, Peter Haufs-Brusberg, and Jens Herder, International Joint Conference on Awareness Science and Technology & Ubi-Media Computing (iCAST 2013 & UMEDIA 2013), Hamamatsu/Aizu-Wakamatsu/Duesseldorf, November, 2013.
Virtuelles Studio



Seitenanfang ^^ 

Kontakt >> 
Suche >> 
Impressum >>