mlsanityData to 7 Oct 2026

Understanding How Image Quality Affects Deep Neural Networks

arXiv 1604.04004 ↗First posted 14 Apr 2016cs.CVcited by 118

Not on the map. Why?

Abstract (arXiv)

Image quality is an important practical challenge that is often overlooked in the design of machine vision systems. Commonly, machine vision systems are trained and tested on high quality image datasets, yet in practical applications the input images can not be assumed to be of high quality. Recently, deep neural networks have obtained state-of-the-art performance on many machine vision tasks. In this paper we provide an evaluation of 4 state-of-the-art deep neural network models for image classification under quality distortions. We consider five types of quality distortions: blur, noise, contrast, JPEG, and JPEG2000 compression. We show that the existing networks are susceptible to these quality distortions, particularly to blur and noise. These results enable future work in developing deep neural networks that are more invariant to quality distortions.

Key facts. Cited by 118 papers in the map (data through 7 Oct 2026). First posted on arXiv 14 Apr 2016.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 22; 2022: 30; 2023: 22; 2024: 17; 2025: 12; 2026: 15.

118 papers.

Sort:

By area:

  1. 2023MD-VQA: Multi-Dimensional Quality Assessment for UGC Live Videoscited by 46
  2. 2021When Human Pose Estimation Meets Robustness: Adversarial Algorithms and Benchmarkscited by 41
  3. 2021PRIME: A few primitives can boost robustness to common corruptionscited by 40
  4. 2023Image Shortcut Squeezing: Countering Perturbative Availability Poisons with Compressioncited by 39
  5. 2022Towards Semantic Communications: Deep Learning-Based Image Semantic Codingcited by 35
  6. 2024Embracing Events and Frames with Hierarchical Feature Refinement Network for Object Detectioncited by 27
  7. 2022Benchmarking the Robustness of Deep Neural Networks to Common Corruptions in Digital Pathologycited by 26
  8. 2022A Survey on Deep Learning for Skin Lesion Segmentationcited by 22
  9. 2023ECG-Image-Kit: A Synthetic Image Generation Toolbox to Facilitate Deep Learning-Based Electrocardiogram Digitizationcited by 20
  10. 2021Certified Adversarial Defenses Meet Out-of-Distribution Corruptions: Benchmarking Robustness and Simple Baselinescited by 17
  11. 2022NoisyMix: Boosting Model Robustness to Common Corruptionscited by 17
  12. 2022Augraphy: A Data Augmentation Library for Document Imagescited by 14
38 more
  1. 2024The METRIC-framework for assessing data quality for trustworthy AI in medicine: a systematic reviewcited by 13
  2. 2021Unravelling the Effect of Image Distortions for Biased Prediction of Pre-trained Face Recognition Modelscited by 11
  3. 2021Anti-aliasing Deep Image Classifiers using Novel Depth Adaptive Blurring and Activation Functioncited by 11
  4. 2021DeepAdversaries: Examining the Robustness of Deep Learning Models for Galaxy Morphology Classificationcited by 10
  5. 2024I-SplitEE: Image classification in Split Computing DNNs with Early Exitscited by 9
  6. 2021Early-exit deep neural networks for distorted images: providing an efficient edge offloadingcited by 8
  7. 2023DeepAstroUDA: Semi-Supervised Universal Domain Adaptation for Cross-Survey Galaxy Morphology Classification and Anomaly Detectioncited by 8
  8. 2021Packet-Loss-Tolerant Split Inference for Delay-Sensitive Deep Learning in Lossy Wireless Networkscited by 7
  9. 2022Perceptual Video Coding for Machines via Satisfied Machine Ratio Modelingcited by 7
  10. 2023Quantifying the LiDAR Sim-to-Real Domain Shift: A Detailed Investigation Using Object Detectors and Analyzing Point Clouds at Target-Levelcited by 7
  11. 2023On the Robustness of Object Detection Models on Aerial Imagescited by 7
  12. 2021AQuA: Analytical Quality Assessment for Optimizing Video Analytics Systemscited by 6
  13. 2021A trainable monogenic ConvNet layer robust in front of large contrast changes in image classificationcited by 6
  14. 2022Lost in Compression: the Impact of Lossy Image Compression on Variable Size Object Detection within Infrared Imagerycited by 6
  15. 2023Confidence-based Reliable Learning under Dual Noisescited by 6
  16. 2021Operationalizing Convolutional Neural Network Architectures for Prohibited Object Detection in X-Ray Imagerycited by 5
  17. 2022Certifying Model Accuracy under Distribution Shiftscited by 5
  18. 2022Does lossy image compression affect racial bias within face recognition?cited by 5
  19. 2022Utilizing Explainable AI for improving the Performance of Neural Networkscited by 5
  20. 2024Boosting Visual Recognition in Real-world Degradations via Unsupervised Feature Enhancement Module with Deep Channel Priorcited by 5
  21. 2024Investigating Calibration and Corruption Robustness of Post-hoc Pruned Perception CNNs: An Image Classification Benchmark Studycited by 5
  22. 2025Robust Asymmetric Heterogeneous Federated Learning with Corrupted Clientscited by 5
  23. 2021Interactive Analysis of CNN Robustnesscited by 4
  24. 2022Does Video Compression Impact Tracking Accuracy?cited by 4
  25. 2023Assessment Framework for Deepfake Detection in Real-world Situationscited by 4
  26. 2024ReLaX-VQA: Residual Fragment and Layer Stack Extraction for Enhancing Video Quality Assessmentcited by 4
  27. 2025Benchmarking Image Perturbations for Testing Automated Driving Assistance Systemscited by 4
  28. 2021Robustness Evaluation of Stacked Generative Adversarial Networks using Metamorphic Testingcited by 3
  29. 2021Benchmarking Robustness of Deep Learning Classifiers Using Two-Factor Perturbationcited by 3
  30. 2021W2WNet: a two-module probabilistic Convolutional Neural Network with embedded data cleansing functionalitycited by 3
  31. 2021In-field high throughput grapevine phenotyping with a consumer-grade depth cameracited by 3
  32. 2021Canonical Face Embeddingscited by 3
  33. 2021A New Image Codec Paradigm for Human and Machine Usescited by 3
  34. 2022A New Approach to Improve Learning-based Deepfake Detection in Realistic Conditionscited by 3
  35. 2022Performance Evaluation of Action Recognition Models on Low Quality Videoscited by 3
  36. 2023Enhancing object detection robustness: A synthetic and natural perturbation approachcited by 3
  37. 2023Leveraging the Edge and Cloud for V2X-Based Real-Time Object Detection in Autonomous Drivingcited by 3
  38. 2024CAPEEN: Image Captioning with Early Exits and Knowledge Distillationcited by 3

50 most cited of 118 All 118 (JSON)