Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

A Data Annotation Process for Human Activity Recognition in Public Places

 
: Cormier, Mickael

:
Fulltext urn:nbn:de:0011-n-6383726 (5.1 MByte PDF)
MD5 Fingerprint: 703ef0bee058a2aa0aedf467f0f2325f
(CC) by
Created on: 28.7.2021


Beyerer, Jürgen (Ed.); Zander, Tim (Ed.):
Joint Workshop of Fraunhofer IOSB and Institute for Anthropomatics, Vision and Fusion Laboratory 2020. Proceedings : 27th to the 31st of July 2020, Karlsruhe
Karlsruhe: KIT Scientific Publishing, 2021 (Karlsruher Schriften zur Anthropomatik 51)
ISBN: 978-3-7315-1091-8
DOI: 10.5445/KSP/1000130397
pp.33-51
Fraunhofer Institute of Optronics, System Technologies and Image Exploitation and Institute for Anthropomatics, Vision and Fusion Laboratory (Joint Workshop) <2020, Karlsruhe>
English
Conference Paper, Electronic Publication
Fraunhofer IOSB ()

Abstract
Behavior analysis of individuals in crowds or groups of people in public places through surveillance cameras gains importance for several different actors. Automatically detecting and understanding pedestrians in real-world uncooperative scenarios is very challenging. Common issues such as limited annotated data, unreliable data and annotation quality, and appropriate use of this data for supervised learning often originate in steps preceding the modeling of specialized neural network architectures. In this report, the necessity and requirements for designing a reliable data annotation process are presented. Some precise ideas for automation through neural networks are discussed in a conceptual manner.

: http://publica.fraunhofer.de/documents/N-638372.html