Record link: 
http://hdl.handle.net/20.500.12708/57799
-
Title: 
Evaluation Study on Semantic Object Labelling in Street Scenes
-
Citation: 
Wittmann, A., Gelautz, M., & Seitner, F. (2019). Evaluation Study on Semantic Object Labelling in Street Scenes. In P. M. Roth, A. Pichler, R. Sablatnig, G. Stübl, & M. Vincze (Eds.), Proceedings of the ARW & OAGM Workshop 2019 (pp. 201–202). Verlag der Technischen Universität Graz. https://doi.org/10.3217/978-3-85125-663-5-45
-
Publisher DOI: 
10.3217/978-3-85125-663-5-45
-
Publication Type: 
Inproceedings - Full-Paper Contribution
en
Authors: 
Wittmann, Andreas 
Gelautz, Margrit 
Seitner, Florian 
-
Organisational Unit: 
E193-01 - Forschungsbereich Computer Vision 
-
Published in: 
Proceedings of the ARW & OAGM Workshop 2019 
-
ISBN: 
978-3-85125-663-5
-
DOI of the book: 
10.3217/978-3-85125-663-5
-
Date (published): 
2019
-
Event name: 
ARW & OAGM Workshop 2019 
-
Event date: 
9-May-2019 - 10-May-2019
-
Event place: 
Steyr, Austria
-
Number of Pages: 
2
-
Publisher: 
Verlag der Technischen Universität Graz
-
Publisher: 
Verlag der Technischen Universität Graz
-
Peer reviewed: 
Yes
-
Keywords: 
semantische Annotierung / Straßenszenen / Kontextbasierte Feature / Texturbasierte Feature / SHCM Feature / Texture-layout-filter Feature
-
Abstract: 
We present a processing pipeline for semantic scene labelling that was developed in view of autonomous driving applications. Our study focuses on two different methods for feature selection - Texture-layout-filter (TLF) and Single Histogram Class Models (SHCM) - whose influence on the performance of a random forest classifier is investigated. In tests on the Cityscapes dataset, we assess the effects of parameter variation and observe an improvement of the Intersection over Union score by 44 percent when substituting the TLF by the computationally more demanding SHCM feature.
en
Project title: 
Combined 3D-Vision and Adaptive Front-Lighting System for Safe Autonomous Driving (FFG - Österr. Forschungsförderungs- gesellschaft mbH) 
-
Research Areas: 
Visual Computing and Human-Centered Technology: 100%
-
Science Branch: 
Informatik
-
Appears in Collections:
Conference Paper

Show full item record

Page view(s)

164
checked on Nov 23, 2023

Google ScholarTM

Check