OD-XAI: Explainable AI-Based Semantic Object Detection for Autonomous Vehicles
Source abstract
Abstract
In recent years, artificial intelligence (AI) has become one of the most prominent fields in autonomous vehicles (AVs). With the help of AI, the stress levels of drivers have been reduced, as most of the work is executed by the AV itself. With the increasing complexity of models, explainable artificial intelligence (XAI) techniques work as handy tools that allow naive people and developers to understand the intricate workings of deep learning models. These techniques can be paralleled to AI to increase their interpretability. One essential task of AVs is to be able to follow the road. This paper attempts to justify how AVs can detect and segment the road on which they are moving using deep learning (DL) models. We trained and compared three semantic segmentation architectures for the task of pixel-wise road detection. Max IoU scores of 0.9459 and 0.9621 were obtained on the train and test set. Such DL algorithms are called “black box models” as they are hard to interpret due to their highly complex structures. Integrating XAI enables us to interpret and comprehend the predictions of these abstract models. We applied various XAI methods and generated explanations for the proposed segmentation model for road detection in AVs.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Explainable Artificial Intelligence (XAI) · Advanced Neural Network Applications · Adversarial Robustness in Machine Learning
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Computer Science and Engineering, Institute of Technology, Nirma University, Ahmedabad 382481, Gujarat, India
Department of Information Management, Asia Eastern University of Science and Technology, New Taipei 22064, Taiwan
Centre for Inter-Disciplinary Research and Innovation, University of Petroleum and Energy Studies, Dehradun 248007, Uttarakhand, India
Funding
Ministry of Science and Technology, Taiwan
References (45)
Show source-provided references
1. (2021, December 14). Road Traffic Accidents. Available online: https://www.worldlifeexpectancy.com/cause-of-death/road-traffic-accidents/by-country/.
2. Exploring inattention and distraction in the SafetyNet Accident Causation Database10.1016/j.aap.2012.03.031 ↗
3. Inattention and distraction in fatal road crashes – Results from in-depth crash investigations in Norway10.1016/j.aap.2019.02.004 ↗
4. Rajasekhar, M.V., and Jaswal, A.K. (2015, January 27–29). Autonomous vehicles: The future of automobiles. Proceedings of the 2015 IEEE International Transportation Electrification Conference (ITEC), Chennai, India.10.1109/ITEC-India.2015.7386874 ↗
5. Fusion of Blockchain and AI for Secure Drone Networking underlying 5G Communications10.1002/ett.4176 ↗
6. A Deep Learning-based Cryptocurrency Price Prediction Scheme for Financial Institutions
7. Mankodiya, H., Obaidat, M.S., Gupta, R., and Tanwar, S. (2021, January 15–17). XAI-AV: Explainable Artificial Intelligence for Trust Management in Autonomous Vehicles. Proceedings of the 2021 International Conference on Communications, Computing, Cybersecurity, and Informatics (CCCI), Beijing, China.10.1109/CCCI52664.2021.9583190 ↗
8. Schorr, C., Goodarzi, P., Chen, F., and Dahmen, T. (2021). Neuroscope: An Explainable AI Toolbox for Semantic Segmentation and Image Classification of Convolutional Neural Nets. Appl. Sci., 11.10.3390/app11052199 ↗
9. (2017, May 03). The Self-Driving Car Market Presents a Strong Opportunity for Samsung. Available online: https://www.businessinsider.com/self-driving-car-market-presents-a-strong-opportunity-for-samsung-2017-5?IR=T.
10. Kaymak, C., and Ucar, A. (2019, January 21–22). Semantic Image Segmentation for Autonomous Driving Using Fully Convolutional Networks. Proceedings of the 2019 International Artificial Intelligence and Data Processing Symposium (IDAP), Malatya, Turkey.10.1109/IDAP.2019.8875923 ↗
11. Cultrera, L., Seidenari, L., Becattini, F., Pala, P., and Del Bimbo, A. (2020, January 14–19). Explaining Autonomous Driving by Learning End-to-End Visual Attention. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, Seattle, WA, USA.10.1109/CVPRW50498.2020.00178 ↗
12. Wang, C., Weisswange, T.H., Kruger, M., and Wiebel-Herboth, C.B. (2021, January 11–15). Human-Vehicle Cooperation on Prediction-Level: Enhancing Automated Driving with Human Foresight. Proceedings of the 2021 IEEE Intelligent Vehicles Symposium Workshops (IV Workshops), Nagoya, Japan.10.1109/IVWorkshops54471.2021.9669247 ↗
13. Kothawade, S., Khandelwal, V., Basu, K., Wang, H., and Gupta, G. (2021). AUTO-DISCERN: Autonomous Driving Using Common Sense Reasoning. arXiv.
14. Intelligent Semantic Segmentation for Self-Driving Vehicles Using Deep Learning10.1155/2022/6390260 ↗
15. (2018, January 12). 10 Charts That Will Change Your Perspective On Artificial Intelligence’s Growth. Available online: https://www.forbes.com/sites/louiscolumbus/2018/01/12/10-charts-that-will-change-your-perspective-on-artificial-intelligences-growth/?sh=570e3c8a4758.
16. Explainable Artificial Intelligence (XAI) to Enhance Trust Management in Intrusion Detection Systems Using Decision Tree Model10.1155/2021/6634811 ↗
17. Can we open the black box of AI10.1038/538020a ↗
18. (2018, May 25). The Importance of Human Interpretable Machine Learning. Available online: https://towardsdatascience.com/human-interpretable-machine-learning-part-1-the-need-and-importance-of-model-interpretation-2ed758f5f476.
19. Mitsuhara, M., Fukui, H., Sakashita, Y., Ogata, T., Hirakawa, T., Yamashita, T., and Fujiyoshi, H. (2019). Embedding Human Knowledge into Deep Neural Network via Attention Map. arXiv.
20. Wang, H., and Abraham, Z. (2015, January 12–17). Concept drift detection for streaming data. Proceedings of the 2015 international joint conference on neural networks (IJCNN), Killarney, Ireland.
21. Coffin, D., Oliver, S., and VerWey, J. (2022, March 02). Building Vehicle Autonomy: Sensors, Semiconductors, Software and US Competitiveness. Available online: https://ssrn.com/abstract=3492778.
22. Lidar for Autonomous Driving: The Principles, Challenges, and Trends for Automotive Lidar and Perception Systems
23. Long, J., Shelhamer, E., and Darrell, T. (2015). Fully Convolutional Networks for Semantic Segmentation. arXiv.10.1109/CVPR.2015.7298965 ↗
24. Mohan, R. (2014). Deep Deconvolutional Networks for Scene Parsing. arXiv.
25. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014). Rich feature hierarchies for accurate object detection and semantic segmentation. arXiv.10.1109/CVPR.2014.81 ↗
26. Ronneberger, O., Fischer, P., and Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. arXiv.10.1007/978-3-319-24574-4_28 ↗
27. Fritsch, J., Kuehnl, T., and Geiger, A. (2013, January 6–9). A New Performance Measure and Evaluation Benchmark for Road Detection Algorithms. Proceedings of the International Conference on Intelligent Transportation Systems (ITSC), Hague, The Netherlands.10.1109/ITSC.2013.6728473 ↗
28. Geiger, A., Lenz, P., and Urtasun, R. (2012, January 16–21). Are we ready for autonomous driving? The KITTI vision benchmark suite. Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, RI, USA.10.1109/CVPR.2012.6248074 ↗
29. Augmented Reality Meets Computer Vision: Efficient Data Generation for Urban Driving Scenes10.1007/s11263-018-1070-x ↗
30. (2015). Deep Residual Learning for Image Recognition. arXiv.
31. (2018, November 27). Residual Blocks—Building Blocks of ResNet. Available online: https://towardsdatascience.com/residual-blocks-building-blocks-of-resnet-fd90ca15d6ec.
32. Veit, A., Wilber, M., and Belongie, S. (2016). Residual Networks Behave Like Ensembles of Relatively Shallow Networks. arXiv.
33. (2021, March 22). KITTI RoadSeg. Available online: https://github.com/robertklee/KITTI-RoadSeg.
34. SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation10.1109/TPAMI.2016.2644615 ↗
35. (2019, February 10). Review: SegNet (Semantic Segmentation). Available online: https://towardsdatascience.com/review-segnet-semantic-segmentation-e66f2e30fb96.
36. Simonyan, K., and Zisserman, A. (2015). Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv.
37. Das, A., and Rad, P. (2020). Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Survey. arXiv.
38. Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)10.1109/ACCESS.2018.2870052 ↗
39. Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization10.1007/s11263-019-01228-7 ↗
40. (2020, May 29). Grad-CAM: Visual Explanations from Deep Networks. Available online: https://glassboxmedicine.com/2020/05/29/grad-cam-visual-explanations-from-deep-networks/.
41. Simonyan, K., Vedaldi, A., and Zisserman, A. (2014). Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps. arXiv.
42. (2018, July 11). What Are Saliency Maps in Deep Learning?. Available online: https://analyticsindiamag.com/what-are-saliency-maps-in-deep-learning/.
43. Cheng, L., Liu, Q., and Ronzhin, A. (2016). Edge Detection Using Convolutional Neural Network. Advances in Neural Networks–ISNN 2016, Springer International Publishing.10.1007/978-3-319-40663-3 ↗
44. Rezatofighi, S.H., Tsoi, N., Gwak, J., Sadeghian, A., Reid, I.D., and Savarese, S. (2019, January 15–20). Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.10.1109/CVPR.2019.00075 ↗
45. Yu, J., Jiang, Y., Wang, Z., Cao, Z., and Huang, T.S. (2016, January 15–19). UnitBox: An Advanced Object Detection Network. Proceedings of the 24th ACM International Conference on Multimedia, Amsterdam, The Netherlands.10.1145/2964284.2967274 ↗