English

Safety Metrics for Semantic Segmentation in Autonomous Driving

Computer Vision and Pattern Recognition 2021-09-28 v2 Machine Learning

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 a different level of safety criticality. The result evaluated on an autonomous driving dataset demonstrates the validity and practicality of our proposed methodology.

Keywords

Cite

@article{arxiv.2105.10142,
  title  = {Safety Metrics for Semantic Segmentation in Autonomous Driving},
  author = {Chih-Hong Cheng and Alois Knoll and Hsuan-Cheng Liao},
  journal= {arXiv preprint arXiv:2105.10142},
  year   = {2021}
}

Comments

Paper accepted at IEEE AI Test'21

R2 v1 2026-06-24T02:19:42.820Z