If a Human Can See It, So Should Your System: Reliability Requirements for Machine Vision Components
Abstract
Machine Vision Components (MVC) are becoming safety-critical. Assuring their quality, including safety, is essential for their successful deployment. Assurance relies on the availability of precisely specified and, ideally, machine-verifiable requirements. MVCs with state-of-the-art performance rely on machine learning (ML) and training data, but largely lack such requirements.
BibTeX
@inproceedings{Hu-al:ICSE22,
author = {Boyue Caroline Hu and
Lina Marsso and
Krzysztof Czarnecki and
Rick Salay and
Huakun Shen and
Marsha Chechik},
title = {If a Human Can See It, So Should Your System: Reliability Requirements for Machine Vision Components},
booktitle = {ICSE},
pages = {1145--1156},
publisher = {{ACM}},
year = {2022},
}