RT Conference Proceedings
T1 Expert-driven Rule-based Refinement of Semantic Segmentation Maps for Autonomous Vehicles
A1 Manibardo, Eric L.
A1 Lana, Ibai
A1 Del Ser, Javier
A1 Carballo, Alexander
A1 Takeda, Kazuya
AB Semantic segmentation aims at assigning labels to every pixel of a given image. In the context of autonomous vehicles, semantic segmentation models should be trained with data collected from the traffic network through which vehicles are expected to circulate. Road regulation, weather conditions, and other context features may differ between regions, making local semantic segmentation datasets extremely valuable. However, the high ground truth annotation costs represent a hindrance to the development of such models. The upsurge of powerful feature learning architectures leaves room for semantic segmentation models trained on an unsupervised fashion. This observation vertebrates the purpose of this work: to produce coarse segmentation maps for scene understanding without the need of annotated data. We depart from an unsupervised model that yields low-quality results. The proposed methodology establishes a set of guidelines for the enhancement of segmentation maps. Obtained results expose an improvement of the segmentation quality thanks to the application of our devised guidelines, paving the way for the automatic generation of semantic segmentation datasets.
PB Institute of Electrical and Electronics Engineers Inc.
SN 9798350346916
YR 2023
FD 2023
LK https://hdl.handle.net/11556/2407
UL https://hdl.handle.net/11556/2407
LA eng
NO Manibardo , E L , Lana , I , Del Ser , J , Carballo , A & Takeda , K 2023 , Expert-driven Rule-based Refinement of Semantic Segmentation Maps for Autonomous Vehicles . in IV 2023 - IEEE Intelligent Vehicles Symposium, Proceedings . IEEE Intelligent Vehicles Symposium, Proceedings , vol. 2023-June , Institute of Electrical and Electronics Engineers Inc. , 34th IEEE Intelligent Vehicles Symposium, IV 2023 , Anchorage , United States , 4/06/23 . https://doi.org/10.1109/IV55152.2023.10186738
NO conference
NO Publisher Copyright: © 2023 IEEE.
NO The authors would like to thank the Basque Government for its funding support through the EMAITEK, ELKARTEK and BIKAINTEK PhD support programs (48AFW22019-00002).
DS TECNALIA Publications
RD 26 jul 2024