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 2,601–2,700 of 6,047
An efficient temporal distortion measure of videos based on spacetime texture
Peng, P., Liao, D., & Li, Z.-N. (2017) · Pattern Recognition · doi:10.1016/j.patcog.2017.04.031 · Cited by 10
Power of tempospatially unified spectral density for perceptual video quality assessment
Aabed, M. A., Kwon, G., & AlRegib, G. (2017) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2017.8019333 · Cited by 10
“Moreover, we compared our metric with commonly used and state of the art full-reference VQA metrics such as MOVIE [6], VQM [16], MS-SSIM [17], VIF [18], VSNR [19] and NQM [20].”
Zhang, Y., Liu, X., Liu, H., & Fan, C. (2019) · Signal Processing Image Communication · doi:10.1016/j.image.2019.07.012 · Cited by 9
A Robustness and Low Bit-Rate Image Compression Network for Underwater Acoustic Communication
Zhuang, M., Luo, Y., Ding, X., et al. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-36711-4_10 · Cited by 9
Image quality assessment model based on multi-feature fusion of energy Internet of Things
Zhu, H. (2020) · Future Generation Computer Systems · doi:10.1016/j.future.2020.05.037 · Cited by 9
Multi-level Feature Fusion Facial Expression Recognition Network
Hu, Q., Wu, C., Chi, J., et al. (2020) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc49329.2020.9164733 · Cited by 9
Huang, Y., Chen, X., & Ding, X.-H. (2015) · Multimedia tools and applications · doi:10.1007/s11042-015-2620-7 · Cited by 8
“The performance of the proposed IQA index is also compared with other 8 state-of- the-art and representative FR-IQA models, including PSNR, UQI [20], SSIM [25], MS-SSIM [23], VIF [17], FSIM [11], GMSD [25].”
MethodSubjective and Objective Assesment of Visual Image Quality Metrics and Still Image Codecs
Richter, T., & Larabi, C. (2008) · DCC · doi:10.1109/dcc.2008.44 · Cited by 9
Neural Video Compression using Spatio-Temporal PriorsPreprint
Liu, H., Chen, T., Lu, M., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1902.07383 · Cited by 9
“Performance comparisons have been carried out with well-known H.264/AVC High Profile (HP) [13] and HEVC MP [1] H.264/AVC HP and HEVC MP, e.g., ≈ 38% BD-Rate [14] improvements against the HEVC MP. Note that distortion measurement used in evaluation is the MS-SSIM [15] presented in decibel (dB) scale.”
MethodLearning to Blindly Assess Image Quality in the Laboratory and WildPreprint
Zhang, W., Ma, K., Zhai, G., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.00516 · Cited by 9
“Our method performs significantly better than all competing models on all six databases, and is close to two full-reference models (MS-SSIM [26] and NLPD [27]).”
MethodUnsupervised Pose Flow Learning for Pose Guided SynthesisPreprint
Zheng, H., Chen, L., Xu, C., & Luo, J. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.13819 · Cited by 11
“To quantitatively evaluate the synthesis results, low-level metrics like Structural Similarity (SSIM) [40], Multi-scale Structural Similarity (MS-SSIM) [41] and perceptual-level metrices like Inception Score (IS) [33] and the Perceptual Image Patch Similarity Distance (LPIPS) [45] are measured on different approaches,…”
MethodUnsupervised Anomaly Detection with Multi-scale Interpolated Gaussian DescriptorsPreprint
Chen, Y., Tian, Y., Pang, G., & Carneiro, G. (2021) · arXiv (Cornell University) · Cited by 9
Yan, Z., Wen, F., Ying, R., Ma, C., & Liu, P. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.02782 · Cited by 52
“In image processing, peak signal to noise rate (PSNR), MSE and SSIM/MS-SSIM (Wang et al., 2003; Zhou Wang et al., 2004) are commonly used distortion measures.”
MethodEfficient Convolutional Neural Networks for Image Processing Applications
Chiapputo, N. (2022) · doi:10.12794/metadc1986970 · Cited by 9
Application of CCC–Schoenberg operators on image resampling
Bosner, T., Crnković, B., & Škifić, J. (2019) · BIT Numerical Mathematics · doi:10.1007/s10543-019-00770-7 · Cited by 8
“We hold the MSSIM results are more relevant because it is a better approximation of an image quality perceived by a human visual system [35,36].”
BackgroundTang, K., Kan, N., Jiang, Y., Li, C., Dai, W., Zou, J., & Xiong, H. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3331487 · Cited by 10
Rethinking Supervision in Document Unwarping: A Self-Consistent Flow-Free Approach
Liu, S., Feng, H., & Zhou, W. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3336068 · Cited by 14
“Subsequently, MSSIM is computed using a 5-level pyramid, adhering to the original MSSIM methodology [69].”
BackgroundMethodInfluentialNawaz, O., Fiedler, M., & Khatibi, S. (2024) · Electronics · doi:10.3390/electronics13020329 · Cited by 9
Hassanaly, R., Brianceau, C., Solal, M., Olivier Colliot, A., & Burgos, N. (2024) · The Journal of Machine Learning for Biomedical Imaging · doi:10.59275/j.melba.2024-b87a · Cited by 24
“We use four metrics that are common in the image synthesis literature (Neˇcasov´a et al., 2022): the mean squared error (MSE), the peak signal-to-noise ratio (PSNR), the structural similarity index (SSIM) (Wang et al., 2004) and the multi-scale structural similarity (MS-SSIM) (Wang et al., 2003) (see details in Append…”
BackgroundSelf-supervised High Dynamic Range Imaging: What Can Be Learned from a Single 8-bit Video?
Banterle, F., Marnerides, D., Bashford‐Rogers, T., & Debattista, K. (2024) · ACM Transactions on Graphics · doi:10.1145/3648570 · Cited by 11
Kim, K., Tun, Y. K., Munir, M. S., et al. (2024) · IEEE Network Operations and Management Symposium (NOMS) · doi:10.1109/noms59830.2024.10575676 · Cited by 8
“…a combination of loss functions, such as mean squared error (MSE), mean absolute error (MAE), multi-scale structural similarity index (MS-SSIM) [12], adversarial loss from the GAN [13], and learned perceptual image patch similarity (LPIPS) [14], which measures not only the restoration of…”
BackgroundColored Point Cloud Quality Assessment Using Complementary Features in 3D and 2D Spaces
Cui, M., Zhang, Y., Fan, C., et al. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3443634 · Cited by 8
“…image is computed here to better capture the differences), s and δ s denote the s th scale of the Laplace pyramid and the corresponding weights [35], respectively, k = 5 is the number of scales, Ψ s ( · ) is the IW-SSIM algorithm. α and β denote the perceptual weights of the depth and texture…”
Deep video compression based on Long-range Temporal Context Learning
Wu, K., Li, Z., Yang, Y., et al. (2024) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2024.104127 · Cited by 8
HNN: Hierarchical Noise-Deinterlace Net Towards Image Denoising
Joshi, A., Akalwadi, N., Mandi, C., Desai, C., Tabib, R. A., Patil, U., & Mudenagudi, U. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00306 · Cited by 10
“Unlike [38] and [29], we propose a weighted combinational loss function, comprising of L 1 loss, MS-SSIM [31], and VGG-16 perceptual loss [20] towards image denoising.”
Low Latency Point Cloud Rendering with Learned Splatting
Hu, Y., Gong, R., Sun, Q., & Wang, Y. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00584 · Cited by 8
“In addition to the original point cloud, we also evaluate these methods in the scenario where the point clouds are lossily compressed for streaming.”
MethodZhihui, G., Nakayama, R., Hizukuri, A., et al. (2024) · Radiological Physics and Technology · doi:10.1007/s12194-024-00851-5 · Cited by 8
“MS-SSIM (Multi Scale-Structural Similarity) [22, 23] in the second term measures the similarity of brightness, contrast, and structure between two images.”
Consistency Guided Diffusion Model with Neural Syntax for Perceptual Image Compression
Kuang, H., Ma, Y., Yang, W., Guo, Z., & Liu, J. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3681336 · Cited by 17
Xie, Y., Hu, J., He, K., et al. (2024) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2024.3507815 · Cited by 8
Imprints: Mitigating Watermark Removal Attacks With Defensive Watermarks
Chen, X., Deng, J., Chen, Y., Li, C., Fang, X., Liu, C., & Xu, W. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3536299 · Cited by 15
“• Structural Similarity Index Measure (SSIM) [44] is a commonly-used metric to measure the similarity between where x , y are two windows, µ x , µ y are the means of x , y . σ 2 x , σ 2 y are the corresponding variances and σ xy is the covariance between x , y . c 1 , c 2 are preset parameters.”
Shen, F., Zhou, L., Kuecuekaytekin, K., Eskandar, G., Liu, Z., Wang, H., & Knoll, A. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3544925 · Cited by 11
“MS-SSIM [42] is a statistical metric to calculate multi-scale structural similarity between two images.”
Jiang, R., Yuan, Z., Wang, H., Fan, Z., Zhang, Y., Liang, N., & Yu, X. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3546375 · Cited by 16
“This study introduces the multiscale structural similarity (MS-SSIM) index to determine the optimal parameters [28].”
Higgins, G., Razavi‐Far, R., Zhang, X., David, A., Ghorbani, A. A., & Ge, T. (2025) · Internet of Things · doi:10.1016/j.iot.2025.101558 · Cited by 14
“For the backwards reconstruction attack the Multi-scale Structured Similarity Index (MS-SSIM) (Wang, Simoncelli and Bovik, 2003) is used, which is an extension of the Structural Similarity Index (SSIM) (Wang, Bovik, Sheikh and Simoncelli, 2004), that given two images, in our case original 𝜙 and reconstructed 𝜃 , it pr…”
InfluentialAdvancing Zero-Shot Digital Human Quality Assessment Through Text-Prompted Evaluation
Zhang, Z., Sun, W., Zhou, Y., Wu, H., Li, C., Min, X., Liu, X., Zhai, G., & Lin, W. (2025) · IEEE Transactions on Image Processing · doi:10.1109/tip.2025.3570597 · Cited by 61
“1) Zero-Shot FR Methods: We consider several classical projection-based FR methods: PSNR, SSIM [64], MS-SSIM [65], and GMSD [66].”
EventSplat: 3D Gaussian Splatting from Moving Event Cameras for Real-time Rendering
Yura, T., Mirzaei, A., & Gilitschenski, I. (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/cvpr52734.2025.02503 · Cited by 25
Human-guided urban form generation using multimodal diffusion models
He, M., Liang, Y., Wang, S., Zheng, Y., Wang, Q., Zhuang, D., Li, T., & Zhao, J. (2025) · Building and Environment · doi:10.1016/j.buildenv.2025.113892 · Cited by 11
GS-QA: Comprehensive Quality Assessment Benchmark for Gaussian Splatting View Synthesis
Martin, P., Rodrigues, A., Ascenso, J. M., & Queluz, M. P. (2025) · International Workshop on Quality of Multimedia Experience · doi:10.1109/qomex65720.2025.11219925 · Cited by 23
“…main categories: i) metrics following a classical approach, including FSIM [27], FVVDP [38], GMSD [28], IW-SSIM [29], MAD [30], MSE-RGB, MS-SSIM [31], NLPD [32], PSNR-HVS [33], PSNR-Y, PSNR-YUV, SSIM [4], VIF [34], and VSI [35]; ii) learning-based metrics that are end-to-end trained, such as…”
Dual-Interaction Spatiotemporal Fusion Network for Agricultural Remote Sensing Imagery
Wang, M., Shang, B., Yao, J. T., et al. (2025) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2025.3631404 · Cited by 8
“Here, we employ the multiscale structural similarity (MS-SSIM) [49] as the L s loss to measure the overall structural similarity between ˆ F 1 and F 1 : Given that the SSIM is inherently bounded and rarely attains a perfect value of 1, training samples that achieve sufficiently high SSIM scores…”
Edge‐based spatial concealment of digital dropout error in degraded archived media
Hoque, M. M., Pimentel, M. L. Q., Hasan, M. M., Ahn, K., Kim, J., & Chae, O. (2014) · Electronics Letters · doi:10.1049/el.2014.1055 · Cited by 8
Automatic prediction of saliency on JPEG distorted images
Mittal, A., Moorthy, A. K., Bovik, A. C., & Cormack, L. K. (2011) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex.2011.6065702 · Cited by 10
Self-citationIncorporating visual attention models into video quality metrics
Akamine, W. Y. L., & Farias, M. C. Q. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2039780 · Cited by 10
“Multi-Scale Structural Similarity Index (MS-SSIM) is a variation of the SSIM metric proposed by Wang et al.17 The method provides more flexibility than single-scale SSIM in incorporating the variations of image resolution and viewing conditions.”
MethodInfluentialFDQM: Fast Quality Metric for Depth Maps Without View Synthesis
Jang, W.-D., Chung, T.-Y., Sim, J.-Y., & Kim, C.-S. (2015) · IEEE transactions on circuits and systems for video technology (Print) · doi:10.1109/tcsvt.2014.2372343 · Cited by 8
“Wang et al. [17] also proposed the multiscale SSIM (MS-SSIM) to compute SSIM at different image scales.”
MethodInfluentialLearning sparse models for image quality assessment
Guha, T., Nezhadarya, E., & Ward, R. (2014) · IEEE International Conference on Acoustics, Speech, and Signal Processing · doi:10.1109/icassp.2014.6853576 · Cited by 10
“To improve the performance of SSIM, multiscale extension [2], wavelet transform-based modification [3], gradient-domain implementation [4] and various pooling strategies [5] have been proposed.”
MethodNo reference quality assessment for screen content images
Fang, Y., Yan, J., Li, L., & Wu, J. (2017) · IEEE International Conference on Multimedia & Expo Workshops (ICMEW) · doi:10.1109/icmew.2017.8026329 · Cited by 9
Divisively Normalized Sparse Coding: Toward Perceptual Visual Signal Representation
Zhang, X., Ma, S., Wang, S., Zhang, J., Sun, H., & Gao, W. (2019) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2019.2899005 · Cited by 9
“To evaluate the performance of the proposed SDS index, we compare it with the popular and state-of-the-art IQAs, including PSNR, VSNR [50], SSIM [52], AS-SSIM [88], MS-SSIM [53], GSM [83], IGM [51], GMSM [89], VSI [90], LTG [91], and HaarPSI [58].”
MethodCross Data Set Performance Consistency of Objective Quality Assessment Methods for Light Fields
Mahmoudpour, S., & Schelkens, P. (2020) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex48832.2020.9123130 · Cited by 8
“Twelve widely-used FR image QA algorithms were considered in our evaluation including PSNR, SSIM [18], MS-SSIM [19], IFC [20], UQI [21], VIF [16], IW-PSNR [15], IW-SSIM [15], GMSD [12], MAD [22], SSRM [17], and FSIM [14].”
MethodThe Generation of Virtual Immunohistochemical Staining Images Based on an Improved Cycle-GAN
Liu, S., Li, X., Zheng, A., Yang, F., Liu, Y., Guan, T., & He, Y. (2021) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-030-66785-6_16 · Cited by 10
Perceptually-inspired super-resolution of compressed videos
Ma, D., Afonso, M., Zhang, A., & Bull, D. (2019) · Bristol Research (University of Bristol) · Cited by 12
A review on high dynamic range (HDR) image quality assessment
Gunawan, I. P., Cloramidina, O., Syafa’ah, S. B., et al. (2021) · International Journal on Smart Sensing and Intelligent Systems · doi:10.21307/ijssis-2021-010 · Cited by 8
“This method combined multi-scale capability of structural similarity measure (SSIM) (Wang et al., 2003) with a measure of naturalness.”
MethodOn Architecture Selection for Linear Inverse Problems with Untrained Neural Networks
Sun, Y., Zhao, H., & Scarlett, J. (2021) · Entropy · doi:10.3390/e23111481 · Cited by 8
“• Following [7], three additional image comparison metrics are considered along with the PSNR, namely, the Visual Information Fidelity (VIF) [41], Structural Similarity Index (SSIM) [42], and Multi-Scale SSIM (MS-SSIM) [43].”
MethodSynthetic data for visual machine learning : A data-centric approach
Tsirikoglou, A. (2022) · Linköping studies in science and technology. Dissertations · doi:10.3384/9789179291754 · Cited by 10
Blind deblurring with fractional-order calculus and local minimal pixel prior
Liu, J., Tan, J., Ge, X., Hu, D., & He, L. (2022) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2022.103645 · Cited by 10
Banding vs. Quality: perceptual impact and objective assessment
Krasula, L., Li, Z., Bampis, C. G., Afonso, M., Miret, N. F., & Solé, J. (2022) · Proceedings - International Conference on Image Processing · doi:10.1109/icip46576.2022.9898041 · Cited by 8
“Table 1 shows the performance of established image and video quality metrics – namely PSNR, SSIM [1], MSSSIM [16], VMAF [2], and CAMBI [3] – on the dataset described in this paper.”
BackgroundThe Performance of Single-Image Super-Resolution Algorithm: EDSR
Lin, K. (2022) · IEEE International Conference on Information Systems and Computer Aided Education (ICISCAE) · doi:10.1109/iciscae55891.2022.9927560 · Cited by 8
“As PSNR or SSIM index increases, the input images resembles better [7-10].”
BackgroundContrast-Controllable Image Enhancement Based on Limited Histogram
Fan, X., Wang, J., Wang, H., et al. (2022) · Electronics · doi:10.3390/electronics11223822 · Cited by 8
SPIDE-Net: Spectral Prior-Based Image Dehazing and Enhancement Network
Qasim, M., & Raja, G. (2022) · IEEE Access · doi:10.1109/access.2022.3221992 · Cited by 8
“Let x and y are the two image patches, extracted from the output and reference images, from the same spatial location, the SSIM [57] can be described as where µ x is the average of x , µ y is the average of y , σ 2 x is the variance of x , σ 2 y is the variance of y , σ xy is the covariance of x…”
BackgroundImpact of Image Compression on In Vitro Cell Migration Analysis
Jalilian, E., Linortner, M., & Uhl, A. (2023) · Computers · doi:10.3390/computers12050098 · Cited by 8
Generating Semantic Adversarial Examples via Feature Manipulation in Latent Space
Wang, S., Chen, S., Chen, T., Nepal, ., Rudolph, C., & Grobler, M. M. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3299408 · Cited by 27
“To satisfy G3, the structural similarity (SSIM) [20] between the original input and the reconstructed instance should exceed a threshold of θ SIM .”
Fast-MFQE: A Fast Approach for Multi-Frame Quality Enhancement on Compressed Video
Chen, K., Chen, J., Zeng, H., & Shen, X. (2023) · Sensors · doi:10.3390/s23167227 · Cited by 12
A Flexible Spatiotemporal Thick Cloud Removal Method with Low Requirements for Reference Images
Zhang, Y., Ji, L., Xunpeng, X. U., et al. (2023) · Remote Sensing · doi:10.3390/rs15174306 · Cited by 8
Spatiotemporal Fusion Model of Remote Sensing Images Combining Single-Band and Multi-Band Prediction
Wang, Z., Fang, S., & Zhang, J. (2023) · Remote Sensing · doi:10.3390/rs15204936 · Cited by 8
Self-citationLuo, J., Zhang, H., Zhuang, Y., Han, L., Chen, K., Hua, Z., Li, C., & Lin, J. (2023) · Sensors · doi:10.3390/s23208614 · Cited by 11
“We utilized three numerical metrics: Fr é chet Inception Distance (FID) [37], Kernel Inception Distance (KID) [38], and multi-scale structural similarity (MS-SSIM) [39] to quantitatively evaluate the performance.”
MethodEnhancing Satisfied User Ratio (SUR) Prediction for VMAF Proxy through Video Quality Metrics
Zhu, J., Amirpour, H., Schatz, R., et al. (2023) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip59821.2023.10402707 · Cited by 8
“• Multi-Scale Structural Similarity Index (MS-SSIM) [25]: Extends SSIM by assessing structural similarity at multiple scales, offering a comprehensive evaluation.”
MethodSemantic Based Generative Compression of Images for Extremely Low Bitrates
Bordin, T., & Maugey, T. (2023) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp59012.2023.10337734 · Cited by 10
“The Mean Squared Error (MSE) naturally comes as a simple and efficient criterion to evaluate the fidelity of the output in terms of distortion [1].”
BackgroundLi, Z., Li, S., Zhang, Z., Fei, W., Wu, F., & Gao, S. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3357784 · Cited by 10
“The m scales of 1 to 5 used for β m and γ m were set to 0.0448, 0.2856, 0.3001, 0.2363, and 0.1333, respectively [53], [54].”
InfluentialAn Evaluation of Quality Metrics for Neural Radiance Field
Onuoha, C., Flaherty, J. A., Luo, S., et al. (2023) · IEEE International Conference on Computational Intelligence and Communication Networks (CICN) · doi:10.1109/cicn59264.2023.10402249 · Cited by 10
“In addition to these metrics, we incorporate traditional full-reference metrics such as Peak Signal-to-Noise Ratio (PSNR) [13], Structural Similarity Index (SSIM) [14], and Multi-scale Structural Similarity Index (MS-SSIM) [15] to provide a comprehensive evaluation of the synthesized NeRF videos.”
MethodSwin Transformer Fusion Network for Image Quality Assessment
Kim, H., & Yim, C. (2024) · IEEE Access · doi:10.1109/access.2024.3378092 · Cited by 10
“Many variants of SSIM have been developed as quality metrics, including multi-scale SSIM (MS-SSIM) [2], complex wavelet SSIM (CW-SSIM) [3], information weighted SSIM (IW-SSIM) [4], feature similarity (FSIM) [5], edge similarity (ESIM) [6], and gradient magnitude similarity deviation (GMSD) [7].”
MethodLearned Video Compression with Spatial-Temporal Optimization
Wang, Y., Huang, Q., Tang, B., et al. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10446198 · Cited by 8
“PSNR and MS-SSIM [22] metrics are used to evaluate the distortion between current and reconstructed frame.”
MethodXu, M., & Lu, F. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i6.28452 · Cited by 8
An Enhanced GAN for Image Generation
Tian, C., Gao, H., Wang, P., & Zhang, B. (2024) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2024.052097 · Cited by 11
Edge-aware interactive refinement network for strip steel surface defects detection
Dong, X., Li, Y., Fu, L., & Liu, J. (2024) · Measurement Science and Technology · doi:10.1088/1361-6501/ad9856 · Cited by 8
“Furthermore, our training approach involves the utilization of a hybrid loss, encompassing BCE loss [44], IOU loss [45], and SSIM loss [46].”
Foveation-based content Adaptive Structural Similarity index
Rimac-Drlje, S., Martinović, G., & Zovko-Cihlar, B. (2011) · International Conference on Systems, Signals and Image Processing · Cited by 16
Synthesis quality prediction model based on distortion intolerance
Ryu, S., Kim, S., & Sohn, K. (2014) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2014.7025117 · Cited by 8
“…The degree of warping distortions |δw| is not directly influenced by the degree of depth distortions |δD|, and rather influenced by spatial gradient of color images C around xo, iii) α and β increase when the degree of depth distortion |δD| and spatial gradient of depth image D around xo are…”
BackgroundRegularization of the structural similarity index based on preservation of edge direction
Peng, P., & Li, Z.-N. (2012) · IEEE International Conference on Systems, Man and Cybernetics · doi:10.1109/icsmc.2012.6378054 · Cited by 9
“the distance from the image plane to the viewer, a multi-scale SSIM approach is proposed in [2].”
BackgroundMethodInfluentialFree-viewpoint video sequences: A new challenge for objective quality metrics
Hanhart, P., Bosc, E., Callet, P. L., & Ebrahimi, T. (2014) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp.2014.6958832 · Cited by 9
“In this study, the performance of the following metrics in predicting image quality of FVV sequences was assessed: 1) PSNR: Peak Signal-to-Noise Ratio, 2) SSIM: Structural Similarity Index [14], 3) MS-SSIM: Multi-Scale Structural Similarity Index [15], 4) VIF: Visual Information Fidelity [16], 5) VIFP: VIF, pixel doma…”
MethodDepth perception of distorted stereoscopic images
Wang, J., Wang, S., & Wang, Z. (2015) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp.2015.7340832 · Cited by 8
Self-citationA color intensity invariant low-level feature optimization framework for image quality assessment
Kottayil, N. K., Cheng, I., Frederic Dufaux, A., & Basu, A. (2016) · Signal Image and Video Processing · doi:10.1007/s11760-016-0873-x · Cited by 9
“Although it shows improvement over IW-SSIM in the TID2008/2013 datasets, MAD (2009) is better than FSIM in the CSIQ and LIVE datasets.”
MethodInfluentialQuality assessment of stereoscopic 3D images based on local and global visual characteristics
Chen, L., & Zhao, J. (2017) · IEEE International Conference on Multimedia & Expo Workshops (ICMEW) · doi:10.1109/icmew.2017.8026266 · Cited by 8
Using multiscale analysis for blind quality assessment of DIBR-synthesized images
Gu, K., Qiao, J., Callet, P. L., Xia, Z., & Lin, W. (2017) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2017.8296380 · Cited by 8
“Following the multiscale analysis in [21], we fuse each similarity map and derive”
BackgroundMethodImage encryption based on double random phase encoding
Liu, Z., Yang, M. L., & Yan, W. Q. (2017) · International Conference on Image and Vision Computing New Zealand (IVCNZ) · doi:10.1109/ivcnz.2017.8402486 · Cited by 11
Towards A Semantic Perceptual Image Metric
Chinen, T. T., Ballé, J., Gu, C., Hwang, S. J., Ioffe, S., Johnston, N., Leung, T., Minnen, D. C., O’Malley, S. M., Rosenberg, C. J., et al. (2018) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2018.8451611 · Cited by 9
“Many existing image metrics such as SSIM [1], MS-SSIM [2], PSNR-HVS [3], PSNR-HVS-M [4], FSIM [5], Butteraugli [6], HaarPSI [7] are designed to model specific documented aspects of the human visual system (HVS), such as contrast or color sensitivity, but are unable to access features that would be able to discern objec…”
BackgroundIEGAN: Multi-Purpose Perceptual Quality Image Enhancement Using Generative Adversarial Network
Ghosh, S., Yang, H., Mukherjee, S. S., et al. (2019) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2019.00009 · Cited by 8
Generating Underwater Images by GANs and Similarity Measurement
Liu, M., Yuan, F., Zhu, Y., & Cheng, E. (2018) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.1109/oceanskobe.2018.8559298 · Cited by 10
Multimedia image quality assessment based on deep feature extraction
Ma, X., & Jiang, X.-H. (2019) · Multimedia tools and applications · doi:10.1007/s11042-019-7571-y · Cited by 9
No-reference artifacts measurements based video quality metric
Vranješ, M., Bajčinovci, V., Grbić, R., & Vajak, D. (2019) · Signal Processing Image Communication · doi:10.1016/j.image.2019.07.015 · Cited by 8
Li, X., Dutta, A., & Richtárik, P. (2017) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2017.217 · Cited by 8
Quality Measurement of Images on Mobile Streaming Interfaces Deployed at Scale
Sinno, Z., Moorthy, A. K., Cock, J. D., Li, Z., & Bovik, A. C. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2939733 · Cited by 9
“We considered simple leading objective picture quality predictors and evaluated their performance on the new database: Peak-Signal to-Noise Ratio (PSNR), PSNR-HVS [37], Structural Similarity Index (SSIM) [9], Multi-Scale Structural Similarity Index (MS-SSIM) [38], Visual Information Fidelity (VIF) [39], Additive Disto…”
MethodInfluentialSelf-citationEnd-to-End Learned ROI Image Compression.
Akutsu, H., & Naruko, T. (2019) · Cited by 8
Wang, M.-H., Xie, W.-Y., Zhang, J., & Qin, J. (2020) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2020.2973733 · Cited by 6
“2) Structural similarity (SSIM) [36] and multiscale SSIM (MS-SSIM) [37] are two most widely used objective metrics to estimate subjective quality.”
MethodLiu, M., Po, L.-M., Xu, X., Cheung, K.-W., Zhao, Y., Lau, K. W., & Zhou, C. S. (2020) · IEEE Access · doi:10.1109/access.2020.2984886 · Cited by 8
Image Quality Assessment Based on Quaternion Singular Value Decomposition
Sang, Q., Yang, Y., Liu, L., Song, X., & Wu, X. (2020) · IEEE Access · doi:10.1109/access.2020.2989312 · Cited by 11
“The multi-scale extension of SSIM (MS-SSIM) [7] produces better results than its single-scale counterpart.”
MethodImage Quality Assessment Using a Combination of Hand-Crafted and Deep Features
Ahmed, N., Asif, H. M. S., & Khalid, H. (2020) · Communications in computer and information science · doi:10.1007/978-981-15-5232-8_51 · Cited by 11
Quantitative Evaluation of Dense Skeletons for Image Compression
Wang, J., Terpstra, M. L., Kosinka, J., et al. (2020) · Information · doi:10.3390/info11050274 · Cited by 8
“As MS-SSIM outperforms the best single-scale SSIM model [27], we consider it next in our work.”
BackgroundMethodAudiovisual quality of live music streaming over mobile networks using MPEG-DASH
Rodrigues, R., Počta, P., Melvin, H., Bernardo, M. V., Pereira, M., & Pinheiro, A. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09047-6 · Cited by 9
“It is also common to adapt Image Quality Metrics, such as PSNR (Peak Signal-to-Noise Ratio) and MS-SSIM (Multi-scale Structural Similarity index) [48] using the average of frame-wise measurements.”
MethodInfluentialResidual Forward-Subtracted U-Shaped Network for Dynamic and Static Image Restoration
Jung, H. M., Kim, B. H., & Kim, M. Y. (2020) · IEEE Access · doi:10.1109/access.2020.3011580 · Cited by 8
“The indicators were MAE, frame-to-frame variation, root-mean-square error (RMSE), PSNR, SSIM [33], and multi-scale SSIM (MS-SSIM) [38], as respectively quantified in (10), (11), (12), (13), (4), (5), and (14): where x is the GT image, and ˆ x is the clean image (prediction image or restored image),…”
MethodA nonlocal Weickert type PDE applied to multi-frame super-resolution
Bella, F. Z. A., Hadri, A., Hakim, A., & Laghrib, A. (2020) · Evolution equations and control theory · doi:10.3934/eect.2020084 · Cited by 8
“In fact, we used two classical metrics: the peak-signal-to-noise ratio (PSNR) and the mean structure similarity (SSIM) [36].”
MethodTiotsop, L. F., Mizdos, T., Uhrina, M., Barkowsky, M., Počta, P., & Masala, E. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09704-w · Cited by 8
“In this work, the PSNR, the SSIM, the VIF [25], the Multi-Scale Structural Similarity Image (MS-SSIM) [28], and the Video Multimethod Assessment Fusion (VMAF) [20] are considered.”
Method3D relative geologic time estimation with deep learning
Bi, Z., Geng, Z., Gao, H., Wu, X., & Li, H. (2020) · Second International Meeting for Applied Geoscience & Energy · doi:10.1190/segam2020-3427388.1 · Cited by 8
Parametric Model for Video Streaming Services With Different Spatial and Temporal Resolutions
Song, J., Yang, F., Zhang, W., & Ma, Z. (2020) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3041489 · Cited by 9
Efficient image structural similarity quality assessment method using image regularised feature
Li, Y., Huang, B., Yang, H., Hou, G., Zhang, P., & Duan, J. (2020) · IET Image Processing · doi:10.1049/iet-ipr.2019.1570 · Cited by 9
“[4, 21] have measured the degree of degradation of the structural information of deformed image to assessment the image, and proposed the theory of structural similarity (SSIM).”
BackgroundEnhancing VMAF through New Feature Integration and Model Combination
Zhang, F., Katsenou, A., Bampis, C. G., Krasula, L., Li, Z., & Bull, D. (2021) · Picture Coding Symposium · doi:10.1109/pcs50896.2021.9477458 · Cited by 12
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.