Paper published in a book (Scientific congresses, symposiums and conference proceedings)
Weakly-Supervised Free Space Estimation through Stochastic Co-Teaching
Robinet, François; Parera, Claudia; Hundt, Christian et al.
2022In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Workshops, 2022
Peer reviewed
 

Files


Full Text
Robinet_Weakly-Supervised_Free_Space_Estimation_Through_Stochastic_Co-Teaching_WACVW_2022_paper.pdf
Publisher postprint (754 kB)
Download

All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
weak supervision; Free space; co-teaching
Abstract :
[en] Free space estimation is an important problem for autonomous robot navigation. Traditional camera-based approaches train a segmentation model using an annotated dataset. The training data needs to capture the wide variety of environments and weather conditions encountered at runtime, making the annotation cost prohibitively high. In this work, we propose a novel approach for obtaining free space estimates from images taken with a single road-facing camera. We rely on a technique that generates weak free space labels without any supervision, which are then used as ground truth to train a segmentation model for free space estimation. Our work differs from prior attempts by explicitly taking label noise into account through the use of Co-Teaching. Since Co-Teaching has traditionally been investigated in classification tasks, we adapt it for segmentation and examine how its parameters affect performances in our experiments. In addition, we propose Stochastic Co-Teaching, which is a novel method to select clean samples that leads to enhanced results. We achieve an IoU of 82.6%, a Precision of 90.9%, and a Recall of 90.3%. Our best model reaches 87% of the IoU, 93% of the Precision, and 93% of the Recall of the equivalent fully-supervised baseline while using no human annotations. To the best of our knowledge, this work is the first to use Co-Teaching to train a free space segmentation model under explicit label noise. Our implementation and trained models are freely available online.
Research center :
Interdisciplinary Centre for Security, Reliability and Trust (SnT) > Services and Data management research group (SEDAN)
Disciplines :
Computer science
Author, co-author :
Robinet, François ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SEDAN
Parera, Claudia ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SEDAN
Hundt, Christian ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT)
Frank, Raphaël ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SEDAN
External co-authors :
no
Language :
English
Title :
Weakly-Supervised Free Space Estimation through Stochastic Co-Teaching
Publication date :
04 January 2022
Event name :
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Workshops 2022
Event place :
United States - Hawaii
Event date :
4-01-2022 to 8-01-2022
Audience :
International
Main work title :
Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Workshops, 2022
Pages :
618-627
Peer reviewed :
Peer reviewed
Focus Area :
Computational Sciences
FnR Project :
FNR13301060 - Machine Learning For Risk Assessment In Semi-autonomous Vehicles, 2018 (01/10/2018-31/08/2022) - François Robinet
Available on ORBilu :
since 04 February 2022

Statistics


Number of views
204 (55 by Unilu)
Number of downloads
71 (16 by Unilu)

Scopus citations®
 
3
Scopus citations®
without self-citations
2

Bibliography


Similar publications



Contact ORBilu