# If a Human Can See It, So Should Your System: Reliability Requirements for Machine Vision Components

arXiv: https://arxiv.org/abs/2202.03930 · HTML: https://mlsanity.com/paper/2202.03930/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 58 references that are in the map.
- First posted on arXiv 8 Feb 2022.

## Builds on (most cited first, 25 of 25)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [Certified Adversarial Robustness via Randomized Smoothing](https://mlsanity.com/paper/1902.02918/) (2019)
- [AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty](https://mlsanity.com/paper/1912.02781/) (2019)
- [RobustBench: a standardized adversarial robustness benchmark](https://mlsanity.com/paper/2010.09670/) (2020)
- [Albumentations: fast and flexible image augmentations](https://mlsanity.com/paper/1809.06839/) (2018)
- [Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Coming](https://mlsanity.com/paper/1907.07484/) (2019)
- [Safety Verification of Deep Neural Networks](https://mlsanity.com/paper/1610.06940/) (2016)
- [Generalisation in humans and deep neural networks](https://mlsanity.com/paper/1808.08750/) (2018)
- [Measuring Neural Net Robustness with Constraints](https://mlsanity.com/paper/1605.07262/) (2016)
- [On the effectiveness of adversarial training against common corruptions](https://mlsanity.com/paper/2103.02325/) (2021)
- [Feature-Guided Black-Box Safety Testing of Deep Neural Networks](https://mlsanity.com/paper/1710.07859/) (2017)
- [Requirements Engineering for Machine Learning: Perspectives from Data Scientists](https://mlsanity.com/paper/1908.04674/) (2019)
- [Towards Verified Artificial Intelligence](https://mlsanity.com/paper/1606.08514/) (2016)
- [Identifying Implementation Bugs in Machine Learning based Image Classifiers using Metamorphic Testing](https://mlsanity.com/paper/1808.05353/) (2018)
- [A simple way to make neural networks robust against diverse image corruptions](https://mlsanity.com/paper/2001.06057/) (2020)
- [RobOT: Robustness-Oriented Testing for Deep Learning Systems](https://mlsanity.com/paper/2102.05913/) (2021)
- [Is Deep Learning Safe for Robot Vision? Adversarial Examples against the iCub Humanoid](https://mlsanity.com/paper/1708.06939/) (2017)
- [Deep Networks Can Resemble Human Feed-forward Vision in Invariant Object Recognition](https://mlsanity.com/paper/1508.03929/) (2015)
- [CIFAR10 to Compare Visual Recognition Performance between Deep Neural Networks and Humans](https://mlsanity.com/paper/1811.07270/) (2018)
- [Using Machine Learning Safely in Automotive Software: An Assessment and Adaption of Software Process Requirements in ISO 26262](https://mlsanity.com/paper/1808.01614/) (2018)
- [Metamorphic Detection of Adversarial Examples in Deep Learning Models With Affine Transformations](https://mlsanity.com/paper/1907.04774/) (2019)

## Built on by (most cited first, 11 of 11)

- [How Mature is Requirements Engineering for AI-based Systems? A Systematic Mapping Study on Practices, Challenges, and Future Research Directions](https://mlsanity.com/paper/2409.07192/) (2024)
- [Automotive Perception Software Development: An Empirical Investigation into Data, Annotation, and Ecosystem Challenges](https://mlsanity.com/paper/2303.05947/) (2023)
- [Assessing Visually-Continuous Corruption Robustness of Neural Networks Relative to Human Performance](https://mlsanity.com/paper/2402.19401/) (2024)
- [RBT4DNN: Requirements-based Testing of Neural Networks](https://mlsanity.com/paper/2504.02737/) (2025)
- [A Safety Assurable Human-Inspired Perception Architecture](https://mlsanity.com/paper/2205.07862/) (2022)
- [A Retrospective on ICSE 2022](https://mlsanity.com/paper/2207.12578/) (2022)
- [Towards Assessing Deep Learning Test Input Generators](https://mlsanity.com/paper/2504.02329/) (2025)
- [Effort-Optimized, Accuracy-Driven Labelling and Validation of Test Inputs for DL Systems: A Mixed-Integer Linear Programming Approach](https://mlsanity.com/paper/2507.04990/) (2025)
- [An Empirical Framework for Evaluating Semantic Preservation Using Hugging Face](https://mlsanity.com/paper/2512.07983/) (2025)
- [Human-AI Divergence in Ego-centric Action Recognition under Spatial and Spatiotemporal Manipulations](https://mlsanity.com/paper/2603.08317/) (2026)
- [From Failure to Alignment: A Requirements Engineering Framework for Machine Learning Systems](https://mlsanity.com/paper/2606.31589/) (2026)

## Abstract (arXiv)

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. In this paper, we address the need for defining machine-verifiable reliability requirements for MVCs against transformations that simulate the full range of realistic and safety-critical changes in the environment. Using human performance as a baseline, we define reliability requirements as: 'if the changes in an image do not affect a human's decision, neither should they affect the MVC's.' To this end, we provide: (1) a class of safety-related image transformations; (2) reliability requirement classes to specify correctness-preservation and prediction-preservation for MVCs; (3) a method to instantiate machine-verifiable requirements from these requirements classes using human performance experiment data; (4) human performance experiment data for image recognition involving eight commonly used transformations, from about 2000 human participants; and (5) a method for automatically checking whether an MVC satisfies our requirements. Further, we show that our reliability requirements are feasible and reusable by evaluating our methods on 13 state-of-the-art pre-trained image classification models. Finally, we demonstrate that our approach detects reliability gaps in MVCs that other existing methods are unable to detect.