Original title:
Pedestrian Detector Domain Shift Robustness Evaluation, And Domain Shift Error Mitigation Proposal
Authors:
Zemčík, Tomáš Document type: Papers
Language:
eng Publisher:
Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií Abstract:
This paper evaluates daytime to nighttime traffic image domain shift on Faster R-CNNand SSD based pedestrian and cyclist detectors. Daytime image trained detectors are applied on anewly compiled nighttime image dataset and their performance is evaluated against detectors trainedon both daytime and nighttime images. Faster R-CNN based detectors proved relatively robust, butstill clearly inferior to the models trained on nighttime images, the SSD based model proved noncompetitive.Approaches to the domain shift deterioration mitigation were proposed and future workoutlined.
Keywords:
ADAS; AV; Cyclist detection; Data augmentation; Domain adaptation; Domain shift; Faster R-CNN,SSD; Object detection; Pedestrian detection Host item entry: Proceedings II of the 27st Conference STUDENT EEICT 2021: Selected papers, ISBN 978-80-214-5943-4
Institution: Brno University of Technology
(web)
Document availability information: Fulltext is available in the Brno University of Technology Digital Library. Original record: http://hdl.handle.net/11012/200837