mlsanityData to 7 Oct 2026

Shift-tolerant Perceptual Similarity Metric

arXiv 2207.13686 ↗First posted 27 Jul 2022cs.CVcited by 44

Not on the map. Why?

Abstract (arXiv)

Existing perceptual similarity metrics assume an image and its reference are well aligned. As a result, these metrics are often sensitive to a small alignment error that is imperceptible to the human eyes. This paper studies the effect of small misalignment, specifically a small shift between the input and reference image, on existing metrics, and accordingly develops a shift-tolerant similarity metric. This paper builds upon LPIPS, a widely used learned perceptual similarity metric, and explores architectural design considerations to make it robust against imperceptible misalignment. Specifically, we study a wide spectrum of neural network elements, such as anti-aliasing filtering, pooling, striding, padding, and skip connection, and discuss their roles in making a robust metric. Based on our studies, we develop a new deep neural network-based perceptual similarity metric. Our experiments show that our metric is tolerant to imperceptible shifts while being consistent with the human similarity judgment.

Key facts. Cited by 44 papers in the map (data through 7 Oct 2026). Builds on 27 of its 40 references that are in the map. First posted on arXiv 27 Jul 2022.

Builds on

27 of 40 references.

References · 27

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  5. 2016Perceptual Losses for Real-Time Style Transfer and Super-Resolutioncited by 2,678
  6. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,898
  7. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  8. 2016SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model sizecited by 1,054
  9. 2020Image Quality Assessment: Unifying Structure and Texture Similaritycited by 792
  10. 2019Making Convolutional Networks Shift-Invariant Againcited by 390
  11. 2016Generative Visual Manipulation on the Natural Image Manifoldcited by 348
  12. 2016Generating Images with Perceptual Similarity Metrics based on Deep Networkscited by 221
15 more
  1. 2017Video Frame Interpolation via Adaptive Separable Convolutioncited by 211
  2. 2018Why do deep convolutional networks generalize so poorly to small image transformations?cited by 198
  3. 2018Spatially Transformed Adversarial Examplescited by 195
  4. 2016EnhanceNet: Single Image Super-Resolution Through Automated Texture Synthesiscited by 177
  5. 2020PIPAL: a Large-Scale Image Quality Assessment Dataset for Perceptual Image Restorationcited by 177
  6. 2020How Much Position Information Do Convolutional Neural Networks Encode?cited by 171
  7. 2018PieAPP: Perceptual Image-Error Assessment through Pairwise Preferencecited by 158
  8. 2020On Translation Invariance in CNNs: Convolutional Layers can Exploit Absolute Spatial Locationcited by 141
  9. 2020Delving Deeper into Anti-aliasing in ConvNetscited by 60
  10. 2019E-LPIPS: Robust Perceptual Image Similarity via Random Transformation Ensemblescited by 44
  11. 2020A Loss Function for Generative Neural Networks Based on Watson's Perceptual Modelcited by 39
  12. 2020An Unsupervised Information-Theoretic Perceptual Quality Metriccited by 28
  13. 2021Impact of Aliasing on Generalization in Deep Convolutional Networkscited by 25
  14. 2015Geodesics of learned representationscited by 14
  15. 2020What Does CNN Shift Invariance Look Like? A Visualization Studycited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 5; 2024: 13; 2025: 17; 2026: 8.

44 papers.

Sort:

By area:

  1. 2025Teaching Large Language Models to Regress Accurate Image Quality Scores using Score Distributioncited by 139
  2. 2023Depicting Beyond Scores: Advancing Image Quality Assessment through Multi-modal Language Modelscited by 104
  3. 2025Q-Insight: Understanding Image Quality via Visual Reinforcement Learningcited by 93
  4. 2024Enhancing Descriptive Image Quality Assessment with A Large-scale Multi-modal Datasetcited by 38
  5. 2023Image Super-Resolution with Text Prompt Diffusioncited by 22
  6. 2024Binarized Diffusion Model for Image Super-Resolutioncited by 18
  7. 2023LipSim: A Provably Robust Perceptual Similarity Metriccited by 15
  8. 2024Multiscale Sliced Wasserstein Distances as Perceptual Color Difference Measurescited by 14
  9. 2024VisualCritic: Making LMMs Perceive Visual Quality Like Humanscited by 13
  10. 2024BiM-VFI: Bidirectional Motion Field-Guided Frame Interpolation for Video with Non-uniform Motionscited by 13
  11. 2022DeepDC: Deep Distance Correlation as a Perceptual Image Quality Evaluatorcited by 10
  12. 2025Adversarially Robust CLIP Models Can Induce Better (Robust) Perceptual Metricscited by 10
32 more
  1. 2023Attacking Perceptual Similarity Metricscited by 8
  2. 2023My3DGen: A Scalable Personalized 3D Generative Modelcited by 8
  3. 2025GS-QA: Comprehensive Quality Assessment Benchmark for Gaussian Splatting View Synthesiscited by 8
  4. 2025AceVFI: A Comprehensive Survey of Advances in Video Frame Interpolationcited by 8
  5. 2025HarmonyIQA: Pioneering Benchmark and Model for Image Harmonization Quality Assessmentcited by 7
  6. 2025Augmenting Perceptual Super-Resolution via Image Quality Predictorscited by 7
  7. 2024Misalignment-Robust Frequency Distribution Loss for Image Transformationcited by 6
  8. 2024Foundation Models Boost Low-Level Perceptual Similarity Metricscited by 6
  9. 2025Hallucination Score: Towards Mitigating Hallucinations in Generative Image Super-Resolutioncited by 5
  10. 2025UARE: A Unified Vision-Language Model for Image Quality Assessment, Restoration, and Enhancementcited by 5
  11. 2024XYScanNet: A State Space Model for Single Image Deblurringcited by 4
  12. 2025Q-MambaIR: Accurate Quantized Mamba for Efficient Image Restorationcited by 4
  13. 2026EditHF-1M: A Million-Scale Rich Human Preference Feedback for Image Editingcited by 4
  14. 2024A New Dataset and Framework for Real-World Blurred Images Super-Resolutioncited by 3
  15. 2024Quality Prediction of AI Generated Images and Videos: Emerging Trends and Opportunitiescited by 3
  16. 2024High-Quality Medical Image Generation from Free-hand Sketchcited by 2
  17. 2025Evaluating quality metrics through the lenses of psychophysical measurements of low-level visioncited by 2
  18. 2025A Picture is Worth a Thousand Prompts? Efficacy of Iterative Human-Driven Prompt Refinement in Image Regeneration Taskscited by 2
  19. 2025DeQA-Doc: Adapting DeQA-Score to Document Image Quality Assessmentcited by 2
  20. 2026How far have we gone in Generative Image Restoration? A study on its capability, limitations and evaluation practicescited by 2
  21. 2024Predicting Satisfied User and Machine Ratio for Compressed Images: A Unified Approachcited by 1
  22. 2025Quality assessment of 3D human animation: Subjective and objective evaluationcited by 1
  23. 2025Non-Aligned Reference Image Quality Assessment for Novel View Synthesiscited by 1
  24. 2025D-PerceptCT: Deep Perceptual Enhancement for Low-Dose CT Imagescited by 1
  25. 2026Fine-Grained Human Pose Editing Assessment via Layer-Selective MLLMscited by 1
  26. 2024Compressed image quality assessment using stackingcited by 0
  27. 2025Toward Better Optimization of Low-Dose CT Enhancement: A Critical Analysis of Loss Functions and Image Quality Assessment Metricscited by 0
  28. 2026When Recovery Matters: The Blind Spot of Surrogate Privacy in MLLM Editingcited by 0
  29. 2026Dash2Sim: Closed-Loop Driving Simulation from in-the-wild Dashcam Videoscited by 0
  30. 2026Towards Standardized Light Field Quality Assessment: Hybrid Subjective Benchmarking and Objective Metric Evaluationcited by 0
  31. 2026The Many Senses of Visual Similarity: A Text-Prompted Image Perceptual Metriccited by 0
  32. 2026JPEG AIC2026: A large-scale dataset for fine-grained assessment of image codingcited by 0