Explaining Bounding Boxes in Deep Object Detectors Using Post Hoc Methods for Autonomous Driving Systems
Author:
Nogueira Caio1ORCID, Fernandes Luís12ORCID, Fernandes João N. D.12ORCID, Cardoso Jaime S.12ORCID
Affiliation:
1. Faculty of Engineering, University of Porto, 4200-465 Porto, Portugal 2. Instituto de Engenharia de Sistemas e Computadores, Tecnologia e Ciência, 4200-465 Porto, Portugal
Abstract
Deep learning has rapidly increased in popularity, leading to the development of perception solutions for autonomous driving. The latter field leverages techniques developed for computer vision in other domains for accomplishing perception tasks such as object detection. However, the black-box nature of deep neural models and the complexity of the autonomous driving context motivates the study of explainability in these models that perform perception tasks. Moreover, this work explores explainable AI techniques for the object detection task in the context of autonomous driving. An extensive and detailed comparison is carried out between gradient-based and perturbation-based methods (e.g., D-RISE). Moreover, several experimental setups are used with different backbone architectures and different datasets to observe the influence of these aspects in the explanations. All the techniques explored consist of saliency methods, making their interpretation and evaluation primarily visual. Nevertheless, numerical assessment methods are also used. Overall, D-RISE and guided backpropagation obtain more localized explanations. However, D-RISE highlights more meaningful regions, providing more human-understandable explanations. To the best of our knowledge, this is the first approach to obtaining explanations focusing on the regression of the bounding box coordinates.
Funder
Portuguese National Funds Innovation and Digital Transition Program ATLAS—Trusted Autonomous Navigation
Subject
Electrical and Electronic Engineering,Biochemistry,Instrumentation,Atomic and Molecular Physics, and Optics,Analytical Chemistry
Reference28 articles.
1. Ren, S., He, K., Girshick, R.B., and Sun, J. (2015). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. arXiv. 2. Redmon, J., and Farhadi, A. (2016). YOLO9000: Better, Faster, Stronger. arXiv. 3. Ilková, V., and Ilka, A. (2017, January 6–9). Legal aspects of autonomous vehicles—An overview. Proceedings of the 2017 21st International Conference on Process Control (PC), Strbske Pleso, Slovakia. 4. Ramon, Y., Farrokhnia, R., Matz, S.C., and Martens, D. (2021). Explainable AI for Psychological Profiling from Behavioral Data: An Application to Big Five Personality Predictions from Financial Transaction Records. Information, 12. 5. Renda, A., Ducange, P., Marcelloni, F., Sabella, D., Filippou, M.C., Nardini, G., Stea, G., Virdis, A., Micheli, D., and Rapone, D. (2022). Federated Learning of Explainable AI Models in 6G Systems: Towards Secure and Automated Vehicle Networking. Information, 13.
|
|