Importance-Aware Semantic Segmentation for Autonomous Driving System

Semantic Segmentation (SS) partitions an image into several coherent semantically meaningful parts, and classifies each part into one of the pre-determined classes. In this paper, we argue that existing SS methods cannot be reliably applied to autonomous driving system as they ignore the different importance levels of distinct classes for safe-driving. For example, pedestrians in the scene are much more important than sky when driving a car, so their segmentations should be as accurate as possible. To incorporate the importance information possessed by various object classes, this paper designs an "Importance-Aware Loss" (IAL) that specifically emphasizes the critical objects for autonomous driving. IAL operates under a hierarchical structure, and the classes with different importance are located in different levels so that they are assigned distinct weights. Furthermore, we derive the forward and backward propagation rules for IAL and apply them to deep neural networks for realizing SS in intelligent driving system. The experiments on CamVid and Cityscapes datasets reveal that by employing the proposed loss function, the existing deep learning models including FCN, SegNet and ENet are able to consistently obtain the improved segmentation results on the pre-defined important classes for safe-driving.

Download Full-text

Failure Detection for Semantic Segmentation on Road Scenes Using Deep Learning

Applied Sciences ◽

10.3390/app11041870 ◽

2021 ◽

Vol 11 (4) ◽

pp. 1870

Author(s):

Junho Song ◽

Woojin Ahn ◽

Sangkyoo Park ◽

Myotaeg Lim

Keyword(s):

Prediction Accuracy ◽

Essential Element ◽

Failure Detection ◽

Ground Truth ◽

Semantic Segmentation ◽

Autonomous Driving ◽

Detection Accuracy ◽

Driving System ◽

Imbalance Data ◽

Autonomous Driving System

Detecting failure cases is an essential element for ensuring the safety self-driving system. Any fault in the system directly leads to an accident. In this paper, we analyze the failure of semantic segmentation, which is crucial for autonomous driving system, and detect the failure cases of the predicted segmentation map by predicting mean intersection of union (mIoU). Furthermore, we design a deep neural network for predicting mIoU of segmentation map without the ground truth and introduce a new loss function for training imbalance data. The proposed method not only predicts the mIoU, but also detects failure cases using the predicted mIoU value. The experimental results on Cityscapes data show our network gives prediction accuracy of 93.21% and failure detection accuracy of 84.8%. It also performs well on a challenging dataset generated from the vertical vehicle camera of the Hyundai Motor Group with 90.51% mIoU prediction accuracy and 83.33% failure detection accuracy.

Download Full-text

Autonomy and Responsibility in Hybrid Systems

10.1093/oso/9780190652951.003.0003 ◽

2017 ◽

Cited By ~ 2

Author(s):

Wulf Loh ◽

Janina Loh

Keyword(s):

Hybrid Systems ◽

Hybrid System ◽

Moral Dilemma ◽

Autonomous Driving ◽

Shared Responsibility ◽

Driving System ◽

Autonomous Cars ◽

Self Driving Cars ◽

Trolley Cases ◽

Autonomous Driving System

In this chapter, we give a brief overview of the traditional notion of responsibility and introduce a concept of distributed responsibility within a responsibility network of engineers, driver, and autonomous driving system. In order to evaluate this concept, we explore the notion of man–machine hybrid systems with regard to self-driving cars and conclude that the unit comprising the car and the operator/driver consists of such a hybrid system that can assume a shared responsibility different from the responsibility of other actors in the responsibility network. Discussing certain moral dilemma situations that are structured much like trolley cases, we deduce that as long as there is something like a driver in autonomous cars as part of the hybrid system, she will have to bear the responsibility for making the morally relevant decisions that are not covered by traffic rules.

Download Full-text