Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Video-to-Video face recognition for low-quality surveillance data

: Herrmann, Christian
: Beyerer, Jürgen

Volltext urn:nbn:de:0072-831689 (41 MByte PDF)
MD5 Fingerprint: 43a9a98f7c5e780c5bb6704f777b6f21
Erstellt am: 10.8.2018

Karlsruhe: KIT Scientific Publishing, 2018, IX, 153 S.
Zugl.: Karlsruhe, Inst. für Technologie (KIT), Diss., 2018
Karlsruher Schriften zur Anthropomatik, 36
ISBN: 978-3-7315-0799-4
ISBN: 3-7315-0799-4
Dissertation, Elektronische Publikation
Fraunhofer IOSB ()

The availability of video data is an opportunity and a challenge for law enforcement agencies. Face recognition methods can play a key role in the automated search for persons in the data. This work targets efficient representations of low-quality face sequences to enable fast and accurate face search. Novel concepts for multi-scale analysis, dataset augmentation, CNN loss function, and sequence description lead to improvements over state-of-the-art methods on surveillance video footage.