Analysis of PSNR, SSIM, LPIPS metrics in the context of human perception of visual similarity
Keywords:
image quality assessment, PSNR, SSIM, LPIPS, human perception, visual distortions, generative models, objective metrics, subjective assessmentAbstract
This paper presents a comprehensive comparative analysis of three well-known image quality assessment (IQA) metrics: PSNR, SSIM, and LPIPS. It explores their basic principles, mathematical foundations, advantages, and limitations, particularly as they relate to human visual perception. The evolution of IQA metrics from simple pixel-by-pixel comparisons (PSNR) to structural approaches (SSIM) and, more recently, to learned perceptual metrics (LPIPS) is discussed. A critical analysis of the effectiveness of each metric in assessing various visual distortions, including noise, blur, and compression artifacts, is presented. Inherent issues in human visual perception, such as the role of semantics, texture, color, and visual artifacts, are explored as fundamental causes of discrepancies between objective metric estimates and subjective human judgments. The paper highlights the “unproven effectiveness” of deep features in LPIPS, and discusses its vulnerabilities, such as adversarial attacks and limitations in global semantic understanding. Finally, it outlines directions for future research aimed at developing more robust, interpretable, and perceptually consistent IQA metrics that can better account for the complexity of the human visual system and the evolving demands of modern image processing and generative artificial intelligence technologies
References
Johnson, J., Alahi, A., & Fei-Fei, L. (2016, September). Perceptual losses for real-time style transfer and super-resolution. In European conference on computer vision (pp. 694-711). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-319-46475-6_43.
Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004). Image quality assessment: from error visibility to structural similarity. IEEE transactions on image processing, 13(4), 600-612. https://doi.org/10.1109/TIP.2003.819861.
Breger, A., Biguri, A., Landman, M. S., Selby, I., Amberg, N., Brunner, E., ... & Schönlieb, C. B. (2025). A study of why we need to reassess full reference image quality assessment with medical images. Journal of Imaging Informatics in Medicine, 1-26. https://doi.org/10.1007/s10278-025-01462-1
Zhang, R., Isola, P., Efros, A. A., Shechtman, E., & Wang, O. (2018). The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 586-595). https://arxiv.org/abs/1801.03924.
Arabboev, M., Begmatov, S., Rikhsivoev, M., Nosirov, K., & Saydiakbarov, S. (2024). A comprehensive review of image super-resolution metrics: classical and AI-based approaches. Acta IMEKO, 13(1), 1-8. https://doi.org/10.21014/actaimeko.v13i1.1679.
Gertsiy, O. (2024). Research on graphic data formats for compact representation and comparison of images. Transport systems and technologies, (43), 173-187. https://doi.org/10.32703/2617-9059-2024-43-14.
Zhang, L., Zhang, L., Mou, X., & Zhang, D. (2011). FSIM: A feature similarity index for image quality assessment. IEEE transactions on Image Processing, 20(8), 2378-2386. https://doi.org/10.1109/TIP.2011.2109730.
Shrestha, B. (2005). Evaluation of JPEG2000 for lossless medical image compression. Mississippi State University Libraries. https://www.gri.msstate.edu/publications/docs/2005/03/4328BijayShrestha_2005.pdf.
Russ, J. C. (2006). The image processing handbook. CRC press.
Gonzalez, R.C., & Woods, R.E. (2017). Digital Image Processing. Pearson, New York, 4 editions.
Singh, G.K., Agarwal, A., & Reddy, N.V. (2023). Comparison of PSNR, SSIM, and LPIPS in medical imaging. In 2023 IEEE 12th International Conference on Computing Communication and Networking Technologies (ICCCNT) (pp. 1–6). https:// www.researchgate. net / figure / LPIPS - SSIM - and - PSNR - Metrics - Comparison - This - table - evaluates – the - quality - of - generated _ tbl 2_382080667.
Wang, Z., Simoncelli, E. P., & Bovik, A. C. (2003, November). Multiscale structural similarity for image quality assessment. In The thrity-seventh asilomar conference on signals, systems & computers, 2003 (Vol. 2, pp. 1398-1402). IEEE. https://doi.org/10.1109/ACSSC.2003.1292216.
Kuzovkin, I., Vicente, R., Petton, M., Lachaux, J. P., Baciu, M., Kahane, P., ... & Aru, J. (2018). Activations of deep convolutional neural networks are aligned with gamma band activity of human visual cortex. Communications biology, 1(1), 107. https://doi.org/10.1038/s42003-018-0110-y.
Zhang, K., Liang, J., Van Gool, L., & Timofte, R. (2021). Designing a practical degradation model for deep blind image super-resolution. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 4791-4800). https://doi.org/10.48550/arXiv.2103.14006.
Zhai, G., & Min, X. (2020). Perceptual image quality assessment: a survey. Science China Information Sciences, 63(11), 211301. https://doi.org/10.1007/s11432-019-2757-1.
Gu, S., Bao, J., Chen, D., & Wen, F. (2020, August). Giqa: Generated image quality assessment. In European conference on computer vision (pp. 369-385). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-58621-8_22.
Shoshan, A., Gandelsman, Yo., Bagon, S, & Dekel, T. (2024). R-LPIPS: An Adversarially Robust Perceptual Similarity Metric. Scientific Reports. https://doi.org/10.48550/arXiv.2307.15157.
Koffka, K. (1935). Principles of Gestalt Psychology. Harcourt, Brace & Co.
Marr, D. (1982). Vision: A computational investigation into the human representation and processing of visual informatio. WH Freeman and Company.
International Telecommunication Union. (2019). Methodology for the subjective assessment of the quality of television pictures (Recommendation BT.500-14). https://www.itu.int/rec/R-REC-BT.500-14-201910-S/en russ
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
