@InProceedings{Manthey2018_400,
<br/>   author = {Robert Manthey and Hussein Hussein and René Erler and Richard Siegel and Danny Kowerko},
<br/>   booktitle = {Studientexte zur Sprachkommunikation: Elektronische Sprachsignalverarbeitung 2018},
<br/>   title = {Realisation of an Audio & Video Laboratory for Precise Object Localization and Tracking},
<br/>   year = {2018},
<br/>   editor = {André Berton and Udo Haiber and Wolfgang Minker},
<br/>   month = mar,
<br/>   pages = {135--142},
<br/>   publisher = {TUDpress, Dresden},
<br/>   abstract  = {This paper presents the realisation of an audiovisual laboratory for detec-
tion, localisation, classification and tracking of objects in indoor environments us-
ing visual as well as audio information. The laboratory is property of the endowed
junior professorship Media Computing at the Chemnitz University of Technology.
Visual information is retrieved by 10 optical embedded smart stereo sensors (In-
tenta S2000). The visual sensors can be used to trigger an identification process of
objects and persons as well as a tracking operation to follow them in the field of
view. For audio signal processing, a total of 64 microphones and 16 loudspeakers
are used. Three microphone arrays are constructed using three different types of
microphones. The camera sensors and microphones can be mounted in different
positions, directions and heights. The loudspeakers can be set up freely within
the tracking area using standard monitor stands. Within a separate air-conditioned
server room, all audio signals are pre-amplified and AD/DA converted using in-
dustry standard rack-mounted audio hardware. A server cluster and workstations
with high-end Nvidia P6000 graphics cards provide the raw processing power for
all processing tasks within the scope of use cases for the laboratory. Software
which is available for users in the laboratory comprises commercial products, such
as Steinberg Cubase 8.5, as well as self-developed solutions, such as an audio &amp;
video localisation and annotation tool. This will be complemented by documenting
acoustic sources (e.g. loudspeakers) and detectors (microphones) into a Blender-
animated 3D laboratory using the video sensor data. Interested individuals will
also have the opportunity to make use of the laboratory&#039;s extra peripheral equip-
ment, e.g. a Yamaha DGX-660 Keyboard, further types of mobile wireless micro-
phones, HTC Wive VR glasses and wireless Internet of Things (IoT) smart home
sensor probes for motion capturing. The active tracking area comprises a total of
82 sensors.},
<br/>   isbn = {978-3-959081-28-3},
<br/>   issn = {0940-6832},
<br/>   keywords = {Poster},
<br/>   url = {https://www.essv.de/pdf/2018_135_142.pdf},
<br/>}