Safety Metrics for Semantic Segmentation in Autonomous Driving

Chih Hong Cheng, Alois Knoll, Hsuan Cheng Liao

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

Within the context of autonomous driving, safety-related metrics for deep neural networks have been widely studied for image classification and object detection. In this paper, we further consider safety-Aware correctness and robustness metrics specialized for semantic segmentation. The novelty of our proposal is to move beyond pixel-level metrics: Given two images with each having $n$ pixels being class-flipped, the designed metrics should, depending on the clustering of pixels being class-flipped or the location of occurrence, reflect different levels of safety criticality. The result evaluated on an autonomous driving dataset demonstrates the validity and practicality of our proposed methodology.

Original languageEnglish
Title of host publicationProceedings - 3rd IEEE International Conference on Artificial Intelligence Testing, AITest 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages57-64
Number of pages8
ISBN (Electronic)9781665434812
DOIs
StatePublished - Aug 2021
Event3rd IEEE International Conference on Artificial Intelligence Testing, AITest 2021 - Virtual, Online, United Kingdom
Duration: 23 Aug 202126 Aug 2021

Publication series

NameProceedings - 3rd IEEE International Conference on Artificial Intelligence Testing, AITest 2021

Conference

Conference3rd IEEE International Conference on Artificial Intelligence Testing, AITest 2021
Country/TerritoryUnited Kingdom
CityVirtual, Online
Period23/08/2126/08/21

Fingerprint

Dive into the research topics of 'Safety Metrics for Semantic Segmentation in Autonomous Driving'. Together they form a unique fingerprint.

Cite this