Multiscale structural similarity for image quality assessment
The DOI matches its Crossref record; 16 of the 16 references it cites were found in a registry or an index; none of the 14 sources its claims cite is checked for support yet.
Abstract
The structural similarity image quality paradigm is based on the assumption that the human visual system is highly adapted for extracting structural information from the scene, and therefore a measure of structural similarity can provide a good approximation to perceived image quality. This paper proposes a multiscale structural similarity method, which supplies more flexibility than previous single-scale methods in incorporating the variations of viewing conditions. We develop an image synthesis method to calibrate the parameters that define the relative importance of different scales. Experimental comparisons demonstrate the effectiveness of the proposed method.
Overview
CoChat Summary
TL;DR
This paper proposes a multi-scale structural similarity (MS-SSIM) image quality method that combines contrast and structure comparisons across scales and luminance comparison at the coarsest scale. On the LIVE JPEG/JPEG2000 database, MS-SSIM outperformed PSNR, the Sarnoff model, and single-scale SSIM variants on all reported evaluation criteria.
Key findings
- MS-SSIM achieved the best reported performance: CC 0.969, ROCC 0.966, MAE 3.86, RMS 4.91, and OR 1.16%, compared with the best single-scale SSIM at M=2 (CC 0.963, ROCC 0.959, MAE 4.21, RMS 5.38, OR 2.62%) and PSNR (CC 0.905, ROCC 0.901, MAE 6.53, RMS 8.45, OR 15.7%).
- Single-scale SSIM performance varied with scale; the best single-scale result was at M=2, and single-scale SSIM tended to give higher scores as scale increased.
- The calibrated cross-scale exponents were β1=γ1=0.0448, β2=γ2=0.2856, β3=γ3=0.3001, β4=γ4=0.2363, and α5=β5=γ5=0.1333.
- The authors conclude that multi-scale SSIM provides a meaningful balance between scales and outperforms the best single-scale SSIM and state-of-the-art image quality metrics.
Methods
- The MS-SSIM system iteratively applies a low-pass filter and downsamples by 2; at each scale it computes contrast and structure comparisons, computes luminance comparison only at Scale M, and combines them using exponents normalized so that ∑γj=1.
- Cross-scale parameters were calibrated by image synthesis: 5 scales and 12 distortion levels (range 2^3 to 2^14) were used with 10 original 64×64 images, producing 600 distorted images; 8 subjects selected images of equal perceived quality across scales at a viewing distance of 32 pixels per degree of visual angle.
- Evaluation used the LIVE database with 344 JPEG and JPEG2000 compressed images (bit rate 0.028 to 3.150 bits/pixel), MOS from 13–25 subjective scores, and compared PSNR, Sarnoff JNDmetrix 8.0, single-scale SSIM for M=1–5, and MS-SSIM using CC, MAE, RMS, OR, and ROCC after logistic mapping.
Limitations
- Stated: the image-synthesis calibration approach is “still rather crude,” and the authors are working on a more systematic approach.
- Stated: CSF cannot be directly used to derive the parameters because it is typically measured at visibility threshold with simplified stimuli, whereas the goal is comparing complex structured images at visible distortion levels.
- Evident from the text: calibration used only 10 original 64×64 grayscale images and 8 subjects, including one author, and evaluation was limited to the LIVE JPEG/JPEG2000 database.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 1 Not valid · 14 Not checked · 18 Own results and interpretations
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It can also be observed that the single-scale model tends to supply higher scores with the increase of scales.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, for every one of the objective evaluation criteria, multi-scale SSIM model outperforms all the other models, including the best single-scale SSIM model, suggesting a meaningful balance between scales.”
- Also stated in: § 5. DISCUSSIONS · p. 4 · hedged
“Experiments show that with an appropriate parameter settings, the multi-scale method outperforms the best single-scale SSIM model as well as state-of-the-art image quality metrics.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 5. DISCUSSIONS · p. 4 · hedged
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In general, the subjects agreed with each other on each image more than they agreed with themselves across different images.”
plainstates an associationn = 8- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The resulting parameters we obtained are β 1 = γ 1 = 0.0448, β 2 = γ 2 = 0.2856, β 3 = γ 3 = 0.3001, β 4 = γ 4 = 0.2363, and α 5 = β 5 = γ 5 = 0.1333, respectively.”
plainn = 8- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“However, the SSIM index algorithm introduced in [5] is a single-scale approach.”
What the sources say
- Not matchedcites [5] [5] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, 'Image quality assessment: From error measurement to structural similarity,' IEEE Trans. Image Processing , vol. 13, Jan. 2004. -
More than one paper fits this citation.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
10 more claims
Create a free account to see all 33 claims and every source verdict.
- It may state two separate claims as one. 8
- Probably not a substantial claim. 13
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“Further, as research in quality assessment has demonstrated, incorporating multiscale information when assessing quality produces QA algorithms that perform better in terms of correlation with human perception [13], [29].”
From No-Reference Image Quality Assessment in the Spatial Domain (2012) · cites this paper as [13]
Citations
Cites
Image quality assessment: from error visibility to structural similarity
Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004) · IEEE Transactions on Image Processing · doi:10.1109/tip.2003.819861 · Cited by 57,892
Self-citationA universal image quality index
Wang, Z., & Bovik, A. C. (2002) · IEEE Signal Processing Letters · doi:10.1109/97.995823 · Cited by 5,757
Self-citationImage quality measures and their performance
Eskicioglu, A. M., & Fisher, P. S. (1995) · IEEE Transactions on Communications · doi:10.1109/26.477498 · Cited by 2,117 · cited 1×
“Traditional perceptual image quality assessment methods are based on a bottom-up approach which attempts to simulate the functionality of the relevant early human visual system (HVS) components.”
BackgroundA Parametric Texture Model Based on Joint Statistics of Complex Wavelet Coefficients
Portilla, J., & Simoncelli, E. P. (2000) · International Journal of Computer Vision · doi:10.1023/a:1026553619983 · Cited by 1,806 · cited 2×
“These synthesis methods provide a powerful and efficient means of testing a model, and have the added benefit that the resulting images suggest improvements that might be made to the model [11].”
MethodSelf-citation(1995) · Medical Entomology and Zoology · Cited by 1,212
Pyramid-based texture analysis/synthesis
Heeger, D. J., & Bergen, J. R. (1995) · Proceedings of the 22nd annual conference on Computer graphics and interactive techniques - SIGGRAPH '95 · doi:10.1145/218380.218446 · Cited by 1,095
Video quality assessment based on structural distortion measurement
Wang, Z., Lu, L., & Bovik, A. C. (2004) · Signal Processing: Image Communication · doi:10.1016/s0923-5965(03)00076-6 · Cited by 1,024
Self-citationImage Quality Assessment: From Error Measurement to Structural Similarity
Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004) · Cited by 919
Self-citationTeo, P. C., & Heeger, D. J. (2002) · Proceedings of 1st International Conference on Image Processing · doi:10.1109/icip.1994.413502 · Cited by 471
Objective video quality assessment
(2003) · Handbook of Video Databases · doi:10.1201/9780203489864-45 · Cited by 359
Research works printed but not linked: 2
- 0, JNDmetrix Technology, Sarnoff Corporation
- 2000, Final report from the video quality experts group on the validation of objective models of video quality assessment, VQEG
Cited by
Items 3,401–3,500 of 6,047
Video quality metrics toolkit: An open source software to assess video quality
Moina-Rivera, W., Aguado, J., & García, M. (2023) · SoftwareX · doi:10.1016/j.softx.2023.101427 · Cited by 11
Frequency Compensated Diffusion Model for Real-Scene DehazingPreprint
Wang, J., Wu, S., Yuan, Z., Tong, Q., & Xu, K. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4573127 · Cited by 6
Image quality metrics fail to accurately represent biological information in fluorescence microscopyPreprint
Gunawan, I., Marsh, R. J., Aggarwal, N., Meijering, E. H. W., Cox, S., Lock, J. G., & Culley, S. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.08.05.668508 · Cited by 6
Generative Probabilistic Entropy Modeling With Conditional Diffusion for Learned Image Compression
Cao, M., Dai, W., Li, S., Li, C., Zou, J., Hu, W., & Xiong, H. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3551780 · Cited by 5
“={ anc , nanc } ; ϕ contains learnable parameters of the factorized entropy model [8] of ˆ z ; λ is the hyperparameter to trade-off distortion and bit-rate; distortion is evaluated with two commonly used metrics, i.e. , mean square error (MSE) and MS-SSIM [63].”
WMFusion: a W-shaped dual encoder and single decoder network for multimodal medical image fusion
Shao, Y. F., Yu, L., & Tang, H. (2025) · Applied Intelligence · doi:10.1007/s10489-025-06477-3 · Cited by 4
Adaptive color-corrected multicolor space enhancement network for underwater image enhancement
Xu, D., Xu, W., Zhou, Y., Shu, X., & Qian, Q. (2025) · Multimedia Systems · doi:10.1007/s00530-025-01867-6 · Cited by 4
“Multiscale structural similarity index measure (MS-SSIM) [49] evaluates the similarity between two images by computing the structural similarity index at multiple scales and calculating a weighted average.”
Rethinking mixture of rain removal via depth-guided adversarial learning
Wang, Y., Yan, X., Niu, Y., Gong, L., Guo, Y., & Wei, M. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.107739 · Cited by 6
Ding, L., Xu, H., Jiang, G., Yu, M., Chen, Y., Luo, T., & Ruan, W. (2025) · Displays · doi:10.1016/j.displa.2025.103125 · Cited by 4
Wang, J., Shen, H., Ke, W., & Liu, X. H. (2025) · IEEE Access · doi:10.1109/access.2025.3586416 · Cited by 8
“To enforce stealthiness from a visual perspective on data distribution (density), we use the Multi-Scale Structural Similarity Index (MS-SSIM) [25].”
iG-6DoF: Model-Free 6DoF Pose Estimation for Unseen Object via Iterative 3D Gaussian Splatting
Cao, T., Luo, F., Qin, J., Jiang, Y., Wang, Y., & Xiao, C. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.00603 · Cited by 10
“Each video sequence is segmented into approximately 200 individual images Metrics.”
Hsu, W., & Huang, S.-H. (2025) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2025.3589209 · Cited by 9
“In addition, the MS-SSIM [45], LPIPS [46], DISTS [44], SCC [47], and SAM [48] are used for auxiliary evaluation.”
Yosifov, M., Fröhler, B., Sijbers, J., et al. (2025) · Journal of Nondestructive Evaluation · doi:10.1007/s10921-025-01262-1 · Cited by 4
Few‐shot learning for highly accelerated 3D time‐of‐flight MRA reconstruction
Li, H., Chiew, M., Dragonu, I., Jezzard, P., & Okell, T. W. (2025) · Magnetic Resonance in Medicine · doi:10.1002/mrm.70072 · Cited by 4
“…performance of the proposed method on experimentally acquired in vivo 3D TOF-MRA data, a combination of quantitative metrics was used, including the peak SNR (PSNR), 47 structural similarity index measure (SSIM), 48 normalized mean square error (NMSE), and the vessel-masked SSIM (VM-SSIM).”
Aqua-Splat: Physically-Informed Sonar-Camera Gaussian Splatting for Underwater 3D Reconstruction
Ling, Z., Feng, Y., Ao, M., Xiao, R., Pan, S., Lu, W., & Hu, L. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3615528 · Cited by 6
“…measures the ratio between the maximum possible power of a signal and the power of corrupting noise; Structural Similarity Index Measure (SSIM) [28], which assesses the structural similarity between two images; and Learned Perceptual Image Patch Similarity (LPIPS) [29], a learned metric that…”
APSNR: Artifact Peak Signal-to-Noise Ratio for Image Quality Assessment
Zhu, L., Zhong, B., & Ma, K. (2025) · IEEE Transactions on Image Processing · doi:10.1109/tip.2025.3623864 · Cited by 5
“Eleven state-of-the-art IQA methods are compared: PSNR [1], SSIM [10], MSSIM [11], EPSNR [5], FSIM [12], PSNR-B [8], ESIM [13], WPSNR [3], DISTS [22], DOSS [26], and PCHSIVS [21].”
Way to Build Native AI-Driven 6G Air Interface: Principles, Roadmap, and Outlook
Zhang, P., Niu, K., Liu, Y., Liang, Z., Ma, N., Xu, X., Xu, W., Sun, M., Liu, Y., Wang, X., et al. (2025) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2025.3636923 · Cited by 9
“The performance is tested using multi-scale structure similarity (MS-SSIM) [116] as an evaluation metric.”
DeepDC: Deep Distance Correlation as a Perceptual Image Quality Evaluator
Zhu, H.-W., Chen, B.-L., Zhu, L., Wang, S.-Q., & Lin, W.-S. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2025.3635025 · Cited by 25
“Wang et al. proposed the structural similarity index (SSIM) [5] to capture perceptually salient structural information, followed by multiscale and translation-invariant extensions [28], [29].”
GIViC: Generative Implicit Video Compression
Gao, G., Teng, S., Peng, T., Zhang, F., & Bull, D. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01612 · Cited by 25
“For each rate point, we calculate the bi-trate (bit/pixel, bpp) and quantitatively assess the quality of lossy reconstructions using PSNR and MS-SSIM [78] in the YUV colorspace.”
Hermann, N., Condor, J., & Didyk, P. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02682 · Cited by 4
“Classical examples of full-reference IQMs include mean-absolute error (MAE), mean-squared error (MSE), peak signal-to-noise ratio (PSNR), SSIM [46], FSIM [59], MS-SSIM [45], and LPIPS [62].”
$${\bf{Micro}}{{\mathbb{S}}}{\bf{plit}}$$: semantic unmixing of fluorescent microscopy data
Ashesh, A., Carrara, F., Zubarev, I., Galinova, V., Croft, M., Pezzotti, M., Gong, D., Casagrande, F., Colombo, E., Giussani, S., et al. (2026) · Nature Methods · doi:10.1038/s41592-026-03082-1 · Cited by 12
Joint Underwater Image Enhancement and Captioning Through Multisupervised Chained Task Learning
Li, H., Li, L., Wang, H., Zhang, W., Liu, J., & Ren, P. (2026) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2026.3710783 · Cited by 22
“1 MS-SSIM [55], and gradient loss [56], to restore color fidelity and structural integrity.”
Kang, H., Dai, D., Zheng, J. D., Liang, Z., Chen, S.-W., & Ding, L. (2023) · Frontiers in Sustainable Food Systems · doi:10.3389/fsufs.2023.1144998 · Cited by 6
“We use the Multi-scale Structural Similarity Index (MS-SSIM; Wang et al., 2003), the Erreur Relative Globale Adimensionnelle de Synthèse (ERGAS; Wald, 2000), and Visual Information Fidelity (VIF; Sheikh and Bovik, 2004) to measure the goodness of the reconstructed features.”
MethodAutomated screening of computed tomography using weakly supervised anomaly detection
Hibi, A., Cusimano, M. D., Bilbily, A., Krishnan, R. G., & Tyrrell, P. N. (2023) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-023-02965-4 · Cited by 5
“Multi-scale SSIM [34], ranging from zero to one, was used as a similarity metric for getting the reconstruction error.”
MethodTowards learning-based image compression for storage on DNA support
Strebel, S., Monnier, N., Lazzarotto, D. N., Testolina, M., & Ebrahimi, T. (2023) · SPIE Applications of Digital Image Processing · doi:10.1117/12.2677356 · Cited by 4
Toward Open-World Text-Driven Face Generation and Manipulation via StyleGAN3
Li, Z., Zhang, Z., Liu, P., Liu, Q., & Sun, X. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3436900 · Cited by 6
Liu, J., Xu, Q. Y., & Su, M. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3440911 · Cited by 5
Slope-embedded ViT-based model for lane line detection under occlusions
Su, Y., Shi, X., Wang, R., Zhang, H., Li, Z., Ti, Y., Song, T., & Puig, V. (2024) · Machine Vision and Applications · doi:10.1007/s00138-024-01621-4 · Cited by 4
“In this study, however, a novel approach is adopted by combining MS_SSIM and L1 [43 ] as the loss function specifically for training the image reconstruction decoder, thereby focusing on the precise elimination of lane line obstructions.”
Feature-Preserving Rate-Distortion Optimization in Image Coding for Machines
Fernández-Menduiña, S., Pavéz, E., & Ortega, A. (2024) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp61759.2024.10743266 · Cited by 5
“For perceptual quality, we include Y-PSNR and Y-MS-SSIM [33].”
MethodDP-ID: Interleaving and Denoising to Improve the Quality of DNA Storage Image
Xu, Q., Ma, Y., Lu, Z., & Bi, K. (2024) · Interdisciplinary Sciences Computational Life Sciences · doi:10.1007/s12539-024-00671-6 · Cited by 4
“A higher MS-SSIM indicates a higher similarity between the original image and the reconstructed image, while a lower value suggests a greater difference between them [18].”
Integrating multi-scale attention network with deep image prior for single image restoration
Arumugam, C. D., & Banothu, B. (2024) · International Journal of Information Technology · doi:10.1007/s41870-024-02315-w · Cited by 4
“The metrics such as PSNR (Peak Signal-to-Noise Ratio), SSIM (Structural Similarity Index), and MS-SSIM (Multi-Scale Structural Similarity Index) [34] are used to validate the model.”
Infrared Image Quality Evaluation: Benchmark Database and Computational Model
Gu, K., Liu, H., Gao, Y.-B., Liu, Y., Qiao, J., Zhai, G., & Zhang, W. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3544722 · Cited by 8
“Second, we modify the information theory-inspired fidelity model for 4 Based on a psychophysical experiment in [25], the psychophysical weights { w 1 ,w 2 ,w 3 ,w 4 ,w 5 } are assigned to be { 0 .”
InfluentialAccelerated EPR Imaging Using Deep Learning Denoising
Canavesi, I., Viswakarma, N., Epel, B., McMillan, A., & Kotecha, M. (2025) · Magnetic Resonance in Medicine · doi:10.1002/mrm.30473 · Cited by 5
Qian, Y., Lee, S.-L., & Wu, S.-T. T. (2025) · Journal of the Society for Information Display · doi:10.1002/jsid.2063 · Cited by 4
TPOT: Topology Preserving Optimal Transport in Retinal Fundus Image Enhancement
Dong, X., Zhu, W., Li, X., et al. (2025) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi60581.2025.10981104 · Cited by 4
“This leads to the following objective function: Here, Cost functions C ( x , G θ ( x )) and C ( y , G θ ( y )) are calculated based on the locally Quasi-Convex Multi-Scale Structural Similarity Index Measure [12, 13].”
MethodParameter Choices in Haarpsi for IQA with Medical Images
Karner, C., Gröhl, J., Selby, I., Babar, J., Beckford, J., Else, T., Sadler, T., Shahipasand, S., Thavakumar, A., Roberts, M., et al. (2024) · IEEE International Symposium on Biomedical Imaging · doi:10.1109/isbi60581.2025.10981227 · Cited by 7
Ge, Z., Huang, Z., Jia, C., et al. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3574389 · Cited by 4
“…LIC methodologies [5], [6], [7], [8], [9], [10] have outperformed traditional standardized codecs [3], [4] in terms of rate-distortion (R-D) performances across multiple quality metrics such as Peak-Signal-to-Noise-Ratio (PSNR) and Multi-scale Structural Similarity Index Metric (MS-SSIM) [11].”
An Image Semantic Communication System Based on Swin Transformer and Generative Adversarial Networks
Liang, W., Ma, N., Shen, L., et al. (2024) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshp64532.2024.11100724 · Cited by 4
“Simulation results demonstrate that the proposed method exhibits significant superiority in MS-SSIM [15] compared to traditional methods and CNN-based joint source-channel coding (JSCC) systems.”
Video Quality Assessment for Resolution Cross-Over in Live Sports
Zhu, J., Chen, Y., Wei, H., Sethuraman, S., & Wu, Y. (2025) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme59968.2025.11209509 · Cited by 4
“We compared various VQMs, including PSNR, SSIM [31], MS-SSIM [32], VMAF [8], P1204.”
MVAW-PCQA: A No-reference Point Cloud Quality Assessment via Multi-View Adaptive Weighting
Hamidi, M., Porcu, S., Floris, A., et al. (2025) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex65720.2025.11219985 · Cited by 4
Universal Rate-Distortion-Classification Representations for Lossy Compression
Nguyen, N. N. N., Nguyen, T., Nguyen, T., & Bose, B. (2025) · IEEE Information Theory Workshop (ITW) · doi:10.1109/itw62417.2025.11240384 · Cited by 6
“However, recent research has demonstrated that minimizing distortion alone is insufficient to yield perceptually convincing reconstructions.”
Generative Enhancement for 3D Medical Images
Zhu, L., Codella, N., Chen, D., Jin, Z., Yuan, L., & Yu, L. (2026) · International Journal of Computer Vision · doi:10.1007/s11263-026-02789-0 · Cited by 22
“To measure the similarity between the ground truth volumes and the generated volumes, we employ two metrics: MS-SSIM [69] and LPIPS [70].”
Recompression effects in iris segmentation
Bergmüller, T., Christopoulos, E., Schnöll, M., & Uhl, A. (2015) · International Conference on Biometrics · doi:10.1109/icb.2015.7139040 · Cited by 4
An efficient demosaicing technique using geometrical information
Maalouf, A., Larabi, M., & Susstrunk, S. E. (2012) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2012.6467478 · Cited by 4
Yang, W.-S., Kung, C.-H., & Kung, C.-M. (2011) · Communications in computer and information science · doi:10.1007/978-3-642-24999-0_70 · Cited by 4
Using region-of-interest for quality evaluation of DIBR-based view synthesis methods
Purica, A. I., Valenzise, G., Pesquet‐Popescu, B., & Frederic Dufaux, A. (2016) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex.2016.7498950 · Cited by 4
“Some of the most popular ones include light field rendering [5], concentric mosaics [6] or lumigraph [7], ii) Methods that use implicit geometry such as pixel correspondences computed with optical flow or any other motion estimation technique [8], iii) and finally, methods that use explicit scene geometry in the form…”
MethodPushing the Limits for View Prediction in Video Coding
Ogniewski, J., & Forssén, P.-E. (2017) · International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP) · doi:10.5220/0006131500680076 · Cited by 4
“In table 1, average PSNR and multiscale SSIM (Wang et al., 2003) values are given for each sequence, for the different configurations.”
MethodInfluentialA fusion-based blind image quality metric for blurred stereoscopic images
Chetouani, A. (2017) · International Conference on Advanced Technologies for Signal and Image Processing · doi:10.1109/atsip.2017.8075530 · Cited by 4
Wang, Z., & Rehman, A. (2019) · SMPTE Motion Imaging Journal · doi:10.5594/jmi.2018.2887288 · Cited by 5
Self-citationA Full-Reference Image Quality Assessment Model Based on Quadratic Gradient Magnitude and LOG Signal
Chen, C., & Mou, X. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-34120-6_57 · Cited by 4
A Center-to-Edge Progression for Equirectangular Projected 360° JPEG Images
Kwon, O., Cho, J., & Choi, S. (2021) · IEEE Access · doi:10.1109/access.2021.3049465 · Cited by 4
“…for color images [13]; 2) structural similarity index metric (SSIM) [14] focusing on the structural image difference; 3) multi-scale SSIM (MS-SSIM) [15] computing the SSIM over different resolutions to represent the quality at different resolutions and viewing conditions; 4) feature-similarity…”
MethodAnalysis of the Perceptual Quality Performance of Different HEVC Coding Tools
Atencia, J. R., López-Granado, O., Malumbres, M. P., Martínez-Rach, M., & Wallendael, G. V. (2021) · IEEE Access · doi:10.1109/access.2021.3062938 · Cited by 4
“The SSIM (Structural Similarity) [9] and the MS-SSIM (Multi-Scale SSIM) [8] metrics are based on the hypothesis that the HVS is highly adapted to extract structural information from the scenes.”
MethodChen, Z., Zhang, S., Zhang, J., Hu, Z., Xue, H., & Xu, M. (2021) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-189632 · Cited by 5
“In this paper we chose two largest databases LIVE and TID2013 data-bases for algorithm validation and comparison, the characteristics of which were listed in Table 1 [26].”
Deep Neural Networks for End-to-End Spatiotemporal Video Quality Prediction and Aggregation
Chen, J., Wang, H., Xu, M., Li, G., & Liu, S. (2021) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme51207.2021.9428209 · Cited by 8
“We compare the performance of our model with the following full reference VQA methods: PSNR, SSIM [19], MS-SSIM [20], ST-RRED [21], ST-MAD [22], VMAF [7], DeepVQA [12] and C3DVQA [14].”
MethodInfluentialQuality Assessment of 2.5D Prints Using 2D Image Quality Metrics
Kadyrova, A., Kitanovski, V., & Pedersen, M. (2021) · Applied Sciences · doi:10.3390/app11167470 · Cited by 4
“The Multi-Scale SSIM (MS-SSIM) is selected because it incorporates variations of image resolution and viewing conditions [52].”
MethodA novel image deblocking approach within a graph framework
Zhou, F., Li, H., Yang, L., et al. (2021) · Digital Signal Processing · doi:10.1016/j.dsp.2021.103270 · Cited by 4
Low-light image enhancement network based on multi-stream information supplement
Yang, Y., Hu, W., Huang, S., Tu, W., & Wan, W. (2022) · Multidimensional Systems and Signal Processing · doi:10.1007/s11045-021-00812-w · Cited by 4
“To avoid blurring or loss of details in the image scale transformation, we utilize a multiscale SSIM (MSSSIM) quality evaluation method (Wang et al., 2003) to construct a structural similarity loss function as one term of the joint loss function.”
MethodDeep-learning-based denoising for photon-counting CT: image domain or projection domain?
Karlsson, H., Moro, V., Eguizabal, A., Eriksson, M., Ågren, A., Åkerström, D., & Persson, M. (2022) · SPIE Medical Imaging: Physics of Medical Imaging · doi:10.1117/12.2612480 · Cited by 5
Görüntü Sıkıştırma Algoritmalarının Performans Analizi İçin Değerlendirme Rehberi
Artuğer, F., & Özkaynak, F. (2022) · International Journal of Pure and Applied Sciences · doi:10.29132/ijpas.1012013 · Cited by 4
Three-Dimensional Virtual Optical Clearing With Cycle-Consistent Generative Adversarial Network
Chen, J., Du, Z., & Si, K. (2022) · Frontiers in Physics · doi:10.3389/fphy.2022.965095 · Cited by 4
Zhao, R., Han, Y., & Zhao, J. H. (2022) · Computational Intelligence and Neuroscience · doi:10.1155/2022/4942420 · Cited by 7
“We leverage Peak Signal-to-Noise Ratio (PSNR), Multiscale Structural Similarity (MS-SSIM) [64], Learned Perceptual Image Patch Similarity (LPIPS) [65], Universal Quality Index (UQI) [66], and Feature Similarity Index Measure (FSIM) [67]. higher PSNR indicates that our method could remove artifacts.”
MethodA genetic algorithm for backlight dimming for HDR displays
Duan, L., Debattista, K., Yue, G., et al. (2022) · The Visual Computer · doi:10.1007/s00371-022-02670-4 · Cited by 5
“The metrics used for comparison were the perceptually uniform (PU) [35] versions of PSNR, structural similarity index measure (SSIM)[37], Multi-Scale SSIM[38], visual information fidelity (VIF) [39], along with HDR-VDP-2.”
MethodEffective low-light image enhancement with multiscale and context learning network
Li, Q., Jiang, B., Bo, X., Yang, C., & Wu, X. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-13830-y · Cited by 4
“(i) MultiScale-Structural Similarity Index (MS-SSIM) [36] is an index that measures the similarity of multiscale structure.”
BackgroundInfluentialQuality evaluation of the JPEG Pleno Holography Call for Proposals response
Prazeres, J., Gilles, A., Muhamad, R. K., et al. (2022) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex55416.2022.9900913 · Cited by 5
“In the following, the performance of the PSNR, SSIM [16], MS-SSIM [17], FSIM [18] and VIFP [19] are reported.”
MethodWu, Y., Feng, S., & Huang, M. (2022) · Geocarto International · doi:10.1080/10106049.2022.2153931 · Cited by 4
Hyperspectral Vnir - Swir Image Registration: Do Not Throw Away Those Overlapping Low Snr Bands
Grillini, F., Thomas, J., & George, S. (2022) · Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS) · doi:10.1109/whispers56178.2022.9955080 · Cited by 4
“Two quality metrics are adapted to suit the evaluation of VNIR-SWIR alignment: the Structural Similarity Index Measure (SSIM) [18] is modified to account only for its contrast and structure components and renamed truncated-SSIM (tSSIM): tSSIM(X,Y ) = 2σXσY + c σ2X + σ 2 Y + c · σXY + c/2 σXσY + c/2 (1) In which X and…”
BackgroundInfluentialSupporting robotic welding of aluminium with a laser line scanner-based trigger definition method
Marco-Rider, J., Tran, T. A., Njaastad, E. B., et al. (2022) · 2022 IEEE 20th International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin51773.2022.9976164 · Cited by 4
SEM Image Quality Assessment Based on Intuitive Morphology and Deep Semantic Features
Wang, H., Li, S., Ding, J., Li, S., Dong, L., & Lu, Z. (2022) · IEEE Access · doi:10.1109/access.2022.3216388 · Cited by 4
Study of Deep Learning-based models for Single Image Super-Resolution
Soufi, O., & Belouadha, F.-Z. (2022) · Revue d intelligence artificielle · doi:10.18280/ria.360616 · Cited by 4
StylePortraitVideo: Editing Portrait Videos with Expression Optimization
Seo, K., Oh, S. W., Lu, J., Lee, J., Kim, S., & Noh, J. (2022) · Computer Graphics Forum · doi:10.1111/cgf.14666 · Cited by 9
Rate-accuracy optimized quantization algorithm based on ROI image coding in power line inspection
Zhang, Z., Jiang, W. T., Zhang, Y., Wang, X., & Yang, J. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15271-7 · Cited by 4
“The conventional image compression paradigms achieve minimal pixel-level distortion under the given bit-rate budget based on human visual characteristics [20, 29] while the compact visual representations are needed to analyze the images in power line inspection.”
Liu, Y., Zhu, Z., Wang, S., et al. (2023) · IEEE Transactions on Microwave Theory and Techniques · doi:10.1109/tmtt.2023.3298203 · Cited by 4
“Next, we add the structural similarity (SSIM) to evaluate the performance of all methods, whose definition can be found in [39].”
MethodEnhancing Image Quality via Robust Noise Filtering Using Redescending M-Estimators
Rendón-Castro, A., Mújica‐Vargas, D., Luna-Álvarez, A., & Kinani, J. M. V. (2023) · Entropy · doi:10.3390/e25081176 · Cited by 4
“The noise suppressing quality was quantified by using the peak signal-to-noise ratio (PSNR ) [52] and structural similarity index (SSIM) [53]; the mean absolute error (MAE) was used to quantify the preservation of the fine details of the image [52].”
MethodInfluentialFLAME-VQA: A Fuzzy Logic-Based Model for High Frame Rate Video Quality Assessment
Mrvelj, Š., & Matulin, M. (2023) · Future Internet · doi:10.3390/fi15090295 · Cited by 4
Image Coding Via Perceptually Inspired Graph Learning
Fernández-Menduiña, S., Pavéz, E., & Ortega, A. (2023) · Proceedings - International Conference on Image Processing · doi:10.1109/icip49359.2023.10222182 · Cited by 4
“As distortion metric, we choose the MS-SSIM since 1) it aligns better with the human visual system [3] and 2) the method in [10] obtains the best results with this quantity.”
BackgroundMethodResultInfluentialZhang, W., Ли, Ж., Qiu, A., Luo, B., Shi, Y., Sun, H., & Li, Y. (2023) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss52108.2023.10282627 · Cited by 5
Pagoda: Privacy Protection for Volumetric Video Streaming through Poisson Diffusion Model
Lu, R., Wei, L., Zhu, S., Hu, C., & Wang, D. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3611946 · Cited by 4
“We investigate the denoising effect, evaluated by SSIM [47] between noisy and denoised images in Fig.”
MethodMedical inter-modality volume-to-volume translation
Chen, J., Huai, Y., & Ma, J. (2023) · Journal of King Saud University - Computer and Information Sciences · doi:10.1016/j.jksuci.2023.101821 · Cited by 4
“Another such metric is MS-SSIM3 (multi-scale structural similarity) index (Wang et al., 2003), which expands on the SSIM index by combining luminance information at the highest resolution level with structure and contrast information at several downsampled resolutions.”
A feasibility analysis of image approximation with image quality assessments
Rahman, M. A., Huq, A., Tunny, S. S., et al. (2023) · IET Image Processing · doi:10.1049/ipr2.12993 · Cited by 4
Principal Component Approximation Network for Image Compression
Zhang, S., Zhao, C., & Basu, A. (2023) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3637490 · Cited by 5
“With the rise of deep learning networks in the computer vision field, many sophisticated meth-ods based on deep learning have been proposed for image and video compression [1–3, 7–9, 11, 13, 14, 17–22, 24–29, 33–36, 38, 39, 41–44, 46, 47, 49–53].”
BackgroundImage reconstruction based on nonlinear diffusion model for limited-angle computed tomography
Zhao, X., Jiang, W., Zhang, X., et al. (2024) · Inverse Problems · doi:10.1088/1361-6420/ad2695 · Cited by 4
“As shown in table 2, we use the peak signal to noise ratio (PSNR) [35] and structural similarity (SSIM) [36] to compare the results quantitatively.”
Fuad, M. M., Faiyaz, A., Arnob, N. M. K., Mridha, M. F., Saha, A. K., & Aung, Z. (2024) · IEEE Access · doi:10.1109/access.2024.3370674 · Cited by 9
“Multi-Scale Structural Similarity Index Measure (MS-SSIM) was used to assess the intra-class diversity of the generated samples compared to the training samples [28].”
MethodEfficient Computational Cost Saving in Video Processing for QoE Estimation
Llorente, Á., Pérez, J., Rodrigo, J. A., Jiménez, D., & Menéndez, J. M. (2024) · IEEE Access · doi:10.1109/access.2024.3373193 · Cited by 4
“Other FR IQA models achieve better correlations with subjective scores and visual perception: PSNR based on HVS [40], SSIM [41], MS-SSIM [42], VSNR [43], MAD [44], VIF [45], FSIM [46] and FMSE [47].”
MethodDeep Learning-Based No-Reference Video Streaming QoE Estimation Using WebRTC Statistics
Sakakibara, K., Ohzahata, S., & Yamamoto, R. (2024) · International Conference on Artificial Intelligence in Information and Communication (ICAIIC) · doi:10.1109/icaiic60209.2024.10463278 · Cited by 4
“It is known to correlate better with subjective assessments than existing FR models, such as PSNR [13], SSIM [14], and MS-SSIM [15].”
BackgroundMasked Feature Compression for Object Detection
Dai, C., Song, T., Jin, Y., Ren, Y., Yang, B., & Song, G. (2024) · Mathematics · doi:10.3390/math12121848 · Cited by 4
“Some of them [11–17] are even better than advanced hand-crafted methods (like the still image coding of VVC [18]) in terms of PSNR (peak signal-to-noise ratio) and MS-SSIM [19].”
BackgroundFull-Reference Video Quality Assessment for User Generated Content Transcoding
Qi, Z., Chen, F., Danier, D., Zhang, F., Xu, X., Liu, S., & Bull, D. (2024) · Picture Coding Symposium · doi:10.1109/pcs60826.2024.10566416 · Cited by 6
“C3DVQA, RankDVQA, VIDEVAL and SimpleVQA are deep learning-based VQA methods, and 0.3752 (-1) 0.8711 (0) 0.6755 (0) MS-SSIM [3] 0.5290 (-1) 0.3677 (-1) 0.8753 (0) 0.6788 (0) LPIPS [7] 0.4491 (-1) 0.3087 (-1) 0.8587 (0) 0.6618 (0) VMAF 0.6.1 [6] 0.4378 (-1) 0.3025 (-1) 0.8957 (0) 0.7127 (0) C3DVQA…”
MethodInfluentialFan, X., Peng, C., Jiang, X., Han, Y., & Hou, L. (2024) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2024.104214 · Cited by 4
Harnessing NFC to Generate Standard Optical Barcodes for NFC-Missing Smartphones
Dai, D., An, Z., Pan, Q., et al. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3420410 · Cited by 4
“For a more quantitative assessment, we employ five established full-reference image quality evaluation algorithms [31], [32], [33], [34] to gauge the image resemblance quantitatively.”
MethodAdaptive Image Downscaling for Rate-Accuracy-Latency Optimization of Task-Target Image Compression
Choi, H., Jeong, S., Kwak, S., Jung, S., & Ko, J. H. (2024) · IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas59952.2024.10595930 · Cited by 4
“Most traditional image/video coding schemes such as JPEG [6], HEVC [7] and VVC [8] are designed to reduce reconstruction distortion according to human visual system and are optimized using pixel field metrics like PSNR or MS-SSIM [9].”
MethodWang, X., Song, Z., Zhu, Y., Wang, S., Zhang, L., Shen, D. G., & Wang, Q. (2024) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi56570.2024.10635169 · Cited by 4
Cheng, N., Liu, Z., Deo, Y., Dou, H., Bi, N., Wu, K., Lin, F., Taylor, Z. A., Ravikumar, N., & Frangi, A. F. (2024) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi56570.2024.10635781 · Cited by 7
“…of DDIM, we calculate Fr ´ echet Inception Distance (FID) scores[13], Fr´echet ResNet Distance (FRD) scores[14], Improved Precision (IP), Improved Recall (IR)[15], multiscale structural similarity metric (MS-SSIM)[16] and Four-Grid-Recursive Structural Similarity Index metric (4-G-R SSIM).”
MethodDetermining Thresholds for Optimal Adaptive Discrete Cosine Transformation
Khanov, A. M., Shulzhenko, A., Ворошилова, А. И., et al. (2024) · Algorithms · doi:10.3390/a17080366 · Cited by 4
TDMD: A Database for Dynamic Color Mesh Quality Assessment Study
Yang, Q., Jung, J., Deschamps, T., et al. (2024) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2024.3451526 · Cited by 4
“The MSU video quality metric benchmark [67] is used to compute the following metrics: PSNR, SSIM [16], MS-SSIM [68], VMAF [17], VQM [69] and 3SSIM [70].”
Pilot-Free Semantic Communication Over Multi-User Mimo Fading Channels
Chen, W., Yang, Q., Yang, Z., Duan, Y., & Zhang, Z. (2024) · Proceedings - International Conference on Image Processing · doi:10.1109/icip51287.2024.10648086 · Cited by 5
NeRVA: Joint Implicit Neural Representations for Videos and Audios
Choudhury, A., Singh, P., & Su, G. (2024) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme57554.2024.10687669 · Cited by 4
“…and frying food and so on from different domains such as human activities, animal To measure the reconstruction quality of video, we use the standard PSNR (in dB) and MS-SSIM [24] metrics (Higher is better) and for audio, we use the perceptual metric FAD [25] quality metric (Lower is better).”
MethodDigital Wireless Semantic Communication Enabled by the Discrete Semantic Knowledge Base and Updating
Shen, Y., Bao, Z., Li, L., et al. (2024) · IEEE/CIC International Conference on Communications in China (ICCC) · doi:10.1109/icccworkshops62562.2024.10693810 · Cited by 4
“It uses the Peak Signal Noise Ratio (PSNR) and Multi-Scale Structural Similarity (MS-SSIM) [12] as evaluation metrics to test the end-to-end communication performance.”
MethodJiao, Q., Hu, W., Hao, G., Cheng, J., Peng, X. H., Liu, Z., & Tan, J. (2024) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-024-02498-w · Cited by 6
Wang, Y., Wan, Z., Qiu, Y., & Wang, Z. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3680588 · Cited by 8
Self-citationDeformable NeRF using Recursively Subdivided Tetrahedra
Qiu, Z., Ren, C., Song, K., et al. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3681019 · Cited by 4
A Feature-Driven Inception Dilated Network for Infrared Image Super-Resolution Reconstruction
Huang, J., Wang, H., Li, Y., & Liu, S. (2024) · Remote Sensing · doi:10.3390/rs16214033 · Cited by 4
Privacy-Preserving Live Video Analytics for Drones via Edge Computing
Nagasubramaniam, P., Wu, C., Sun, Y., Karamchandani, N., Zhu, S., & He, Y. (2024) · Applied Sciences · doi:10.3390/app142210254 · Cited by 5
Metadata
- Authors
- Wang, Z., Simoncelli, E. P., & Bovik, A. C.
- Affiliations
- Courant Institute of Mathematical Sciences (US)
- The University of Texas at Austin (US)
- Venue
- The Thrity-Seventh Asilomar Conference on Signals, Systems & Computers, 2003
- Publisher
- IEEE
- Published
- 2004-07-08
- Pages
- 1398-1402
- ISSN
- 1058-6393 (ISSN-L)
- OpenAlex
- W1580389772
- Cited by
- 6,253
- References
- 15
- Citations by year
- 2026: 344 · 2025: 675 · 2024: 729 · 2023: 668 · 2022: 662 · 2021: 641 · 2020: 563 · 2019: 438 · 2018: 305 · 2017: 267 · 2016: 206 · 2015: 173 · 2014: 166 · 2013: 112 · 2012: 89
- FWCI
- 3.98
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Asilomar Conference on Signals, Systems and Computers (ACSSC) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Video Quality Assessment
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image Retrieval and Classification Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Keywords
- structural similarity, image quality assessment, multiscale analysis, human visual system, perceptual image quality, image generation, parameter calibration, viewing conditions
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.