Main content



Loading wiki pages...

Wiki Version:
# The RoboCup Rescue Victim Dataset In this paper, we introduce a victim dataset for the RoboCup Rescue competitions. The RoboCup Rescue robots have to collect points within several disciplines, e.g. a search task within an area to survey simulated baby doll (victim). When a robot comes across a victim, a heat detector does not completely proof if this is a living being and not just a heat emitting somewhat else. Further investigations are necessary so that a face detection could prove the existence of a victim. Lots of face detection approaches can be found in literature, which manly are used for human face recognition. These cannot be straightforward used for victim faces which are, in case of the RoboCup Rescue competitions, typically dolls. Thus we present the results of standard approaches and developed an own approach via bag-of-visual-words (BoVW). Index Terms — dataset, robocup rescue, victim detection, applied computer vision #### Citation `@INPROCEEDINGS{plorenz18, author={P. Lorenz and G. Steinbauer}, booktitle={2018 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)}, title={The RoboCup Rescue Victim Dataset}, year={2018}, volume={}, number={}, pages={1-6}, keywords={Detectors;Face;Support vector machines;Task analysis;Face detection;Standards;dataset;robocup rescue;victim detection;applied computer vision}, doi={10.1109/SSRR.2018.8468605}, ISSN={2475-8426}, month={Aug}, url={} }` #### Acknowledgement I would like to thank the RoboCup Rescue Team [1][2] of the Graz University of Technolgy who helped me a alot for this Bachelor thesis. #### References [1] [2]
OSF does not support the use of Internet Explorer. For optimal performance, please switch to another browser.
This website relies on cookies to help provide a better user experience. By clicking Accept or continuing to use the site, you agree. For more information, see our Privacy Policy and information on cookie use.

Start managing your projects on the OSF today.

Free and easy to use, the Open Science Framework supports the entire research lifecycle: planning, execution, reporting, archiving, and discovery.