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.
Multiscale structural similarity for image quality assessment
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. -
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
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
A 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
Self-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 881
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 3,501–3,600 of 6,047
ASOD: Attention-Based Salient Object Detector for Strip Steel Surface Defects
Yue, H., Li, X., Sun, Y., Zhang, L., Feng, Y., & Guo, H. (2025) · Electronics · doi:10.3390/electronics14050831 · Cited by 5
Gao, L., Liu, B., Fu, P., Xu, M., Zhang, Y., & Huang, Y. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3547529 · Cited by 7
“Restrictions apply. where p denotes the p th pixel, N is the number of pixels, and f iR → D and GT iD are the depth estimation map and depth map ground truth of the i th stage, respectively, where ssim i rgb is the SSIM [53] of depth estimation.”
Labib, S. M. F. R., Mondal, J. J., Manab, M. A., Xiao, X., & Newaz, S. (2025) · Scientific Reports · doi:10.1038/s41598-025-87405-w · Cited by 8
Yan, P., Ward, R. K., Tang, Q., et al. (2025) · IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv61041.2025.00037 · Cited by 4
GLIC: General Format Learned Image Compression
Zhou, M., & Kong, M. (2025) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v39i10.33175 · Cited by 5
Lao, Z., & Yu, H. (2025) · IEEE Conference on Artificial Intelligence (CAI) · doi:10.1109/cai64502.2025.00285 · Cited by 5
“The traditional IQA module evaluates the quality of the image to be assessed using industry-standard formulas, including PSNR, SSIM, and MS-SSIM [30]–[32], [54].”
InfluentialMulti-Domain Spatial-Temporal Redundancy Mining for Efficient Learned Video Compression
Feng, Y., Pan, Z., Lei, J., et al. (2025) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2025.3587532 · Cited by 4
“…D mei and D reci denote the distortion terms for motion estimation and frame reconstruction, which are calculated using MSE or MS-SSIM [52]; λ serves as the Lagrangian multiplier to balance the trade-off between R and D ; ω i is a hierarchical quality factor to alleviate accumulated…”
Ruff, C., Hauser, T., Roder, C., Feucht, D., Bombach, P., Zerweck, L., Staber, D., Paulsen, F., Ernemann, U., & Gohla, G. (2025) · Diagnostics · doi:10.3390/diagnostics15151982 · Cited by 7
Image quality assessment based on multiscale fuzzy gradient similarity deviation
Guo, S., Xiang, T., & Li, X. (2015) · Soft Computing · doi:10.1007/s00500-015-1844-9 · Cited by 5
“We compare the experimental results of our proposed MFGSD with many well-known existing FR-IQA schemes, including PSNR, SSIM (Wang et al. 2004a), MS-SSIM (Wang et al. 2004b), IFC (Sheikh et al. 2005), VIF (Sheikh andBovik2006),VSNR(ChandlerandHemami2007),GSM (Zhu and Wang 2012), and GMSD (Xue et…”
BackgroundMethodResultInfluentialImage quality assessment and color difference
Temel, D., & AlRegib, G. (2014) · IEEE Global Conference on Signal and Information Processing · doi:10.1109/globalsip.2014.7032265 · Cited by 4
On the JPEG 2000 ultrafast mode
Richter, T., & Simon, S. (2012) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2012.6467406 · Cited by 4
“Figure 3 shows the decoding times over the quality measured in multiscale SSIM [8], where the ordinate denotes 10 log(1−SSIM) instead of linear SSIM; because SSIM tends to one for increasing image quality, the logarithm tends to infinity and the resulting quality index behaves similar to PSNR 1 .”
MethodInfluentialPerformance analysis of MC-CDMA system when image transmission is involved
Crăciunescu, R., Voicu, C., Vulpe, A., et al. (2014) · International Conference on Communications (COMM) · doi:10.1109/iccomm.2014.6866742 · Cited by 4
Hou, X., Jiang, G., Ji, R., et al. (2011) · IET Image Processing · doi:10.1049/iet-ipr.2010.0059 · Cited by 4
Efficient video quality assessment based on spacetime texture representation
Peng, P., Cannons, K. J., & Li, Z.-N. (2013) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/2502081.2502168 · Cited by 4
“As in [3], the filter responses are point-wise rectified and summed over a spatiotemporal neighborhood (a spatiotemporal region Π) to yield a measurement of signal energy for this region at each orientation θ̂: Eθ̂(x, y, t) = ∑ (x,y,t)∈Π (G3 θ̂ ∗ V )2 (1) where V = V (x, y, t) denotes the input…”
BackgroundSmart selective encryption of H.264/AVC videos using confidentiality metrics
Dubois, L., Puech, W., & Blanc-Talon, J. (2014) · Annals of Telecommunications · doi:10.1007/s12243-013-0413-4 · Cited by 4
HLFSIM: Objective image quality metric based on ROI analysis
Dostal, P., Krasula, L., & Klima, M. (2012) · IEEE International Carnahan Conference on Security Technology (ICCST) · doi:10.1109/ccst.2012.6393587 · Cited by 4
Supporting binocular visual quality prediction using machine learning
Wang, S., Shao, F., & Jiang, G. (2014) · IEEE International Conference on Multimedia and Expo workshops · doi:10.1109/icmew.2014.6890547 · Cited by 4
“Numerous 2D image quality assessment (2D-IQA) approaches has been proposed over the last several decades, such as Structural SIMilarity (SSIM) [3], multi-scale SSIM (MS-SSIM) [4], UQI (universal quality index) [5], VIF (visual information fidelity) [6], etc.”
BackgroundImage quality assessment based on structural saliency
Zhang, Z., Zhang, J., Wang, X., Guan, Q., & Chen, S. (2014) · IEEE International Conference on Digital Signal Processing (DSP) · doi:10.1109/icdsp.2014.6900714 · Cited by 5
A Novel Approach towards Video Compression for Mobile Internet using Transform Domain Technique
Bhojani, D. R., & Dwivedi, V. V. (2012) · International Journal of Computer Applications · doi:10.5120/9320-3552 · Cited by 4
Perceptual quality assessment of SAR image compression
Hu, A., Zhang, R., Yin, D., et al. (2013) · International Journal of Remote Sensing · doi:10.1080/01431161.2013.846488 · Cited by 4
“IC-SSIM, MS-SSIM, and PIQ employ many HVS characteristics and have a higher correlation with subjective perception than MSSIM.”
MethodInfluentialStereoscopic image quality assessment model with three-component weighted structure similarity
Mao, X., Yu, M., Wang, X., et al. (2010) · International Conference on Audio, Language and Image Processing · doi:10.1109/icalip.2010.5685199 · Cited by 4
Temporal analysis and perceptual weighting for objective video quality measurement
Cardoso, J. V. D. M., de Alencar, M. S., Regis, C. D. M., & Oliveira, Í. P. (2014) · IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI) · doi:10.1109/ssiai.2014.6806028 · Cited by 4
Beyond Standard Noise Models: Evaluating Denoising Algorithms with Respect to Realistic Camera Noise
Seybold, T., Knopp, M., Keimel, C., & Stechele, W. (2014) · International Journal of Semantic Computing · doi:10.1142/s1793351x14400029 · Cited by 5
“The MSSIM better matches the subjective quality, as it shows a lower value for the noise after demosaicking, see the MSSIM in Fig.”
MethodInfluentialImage Denoising Based on Wavelet for IR Images Corrupted by Gaussian, Poisson & Impulse Noises
Islam, M. R., Huang, X., Liao, M., et al. (2013) · Cited by 4
A Novel Image Quality Assessment Method Based on Vision Attention
Lan, Z., Li, Y., & Liu, X. (2015) · IEEE International Conference on Computer and Information Technology (CIT) · doi:10.1109/cit/iucc/dasc/picom.2015.315 · Cited by 5
“Variance, which stands for the structure information of image, can be used to express visual attention; it directly decides how much attention the human visual system (HVS) will distributes to the region.”
BackgroundWatermarking of radiographic images based on quantum noise modulation
Karasad, M., Bouslimi, D., Cozic, M., et al. (2016) · IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI ...) · doi:10.1109/bhi.2016.7455822 · Cited by 4
Efficient Video Quality Assessment via 3D-Gradient Similarity Deviation
Jia, C., He, L., Lu, W. B., Hao, L., & Gao, X. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-23989-7_50 · Cited by 4
No-Reference Image Quality Assessment Based on HVS
Fu, Y., & Wang, S. (2016) · International Symposium on Computer, Consumer and Control · doi:10.1109/is3c.2016.275 · Cited by 4
“SSIM [3], MS-SSIM [4] and VSNR [5] are examples of FR-IQA algorithms.”
MethodA novel structural variation detection strategy for image quality assessment
Zhan, Y., & Zhang, R. (2016) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2016.7532723 · Cited by 4
Playout Continuity Driven Framework for HTTP Adaptive Streaming Over LTE Networks
Chen, Y., & Liu, G. (2016) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2016.2615464 · Cited by 4
No-reference image quality assessment based on local binary patterns
Ненахов, И. Г., Khryashchev, V. V., & Priorov, A. L. (2016) · IEEE East-West Design & Test Symposium (EWDTS) · doi:10.1109/ewdts.2016.7807685 · Cited by 4
Annotation of the medical images using quick response codes
Kazakevičiūtė-Januškevičienė, G., Ušinskas, A., Januškevičius, E., et al. (2017) · Computer Science and Information Systems · doi:10.2298/csis151217044k · Cited by 4
Optimized three scores combination for image quality assessment
Ishiyama, K., Sugiura, Y., & Shimamura, T. (2016) · IEEE Asia Pacific Conference on Circuits and Systems (APCCAS) · doi:10.1109/apccas.2016.7803881 · Cited by 4
Sampled efficient full-reference image quality assessment models
Bampis, C. G., Goodall, T., & Bovik, A. C. (2016) · Asilomar Conference on Signals, Systems and Computers (ACSSC) · doi:10.1109/acssc.2016.7869104 · Cited by 4
Self-citationAmor, M. B., Kammoun, F., & Masmoudi, N. (2017) · Multimedia Tools and Applications · doi:10.1007/s11042-017-5393-3 · Cited by 4
“Unsurprisingly, our metrics with pre-processing remains lower in terms of correlation of the metrics MOVIE, MSSIM and FSIM.”
InfluentialThanki, R. M., Dwivedi, V. V., & Borisagar, K. R. (2018) · Advances in intelligent systems and computing · doi:10.1007/978-981-10-6977-2_1 · Cited by 4
Spatial resolution enhancement for pushbroom-based microscopic hyperspectral imaging
Jiang, Z., Yu, Z., Yu, Y., Huang, Z., Ren, Q., & Li, C. (2019) · Applied Optics · doi:10.1364/ao.58.000850 · Cited by 5
Machine Learning Approaches for Predicting the 10.7 cm Radio Flux from Solar Magnetogram Data
Valdés, J. J., Nikolić, L., & Tapping, K. (2019) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2019.8852332 · Cited by 5
Ermeydan, E. Ş., Çankaya, İ., & Şahın, A. B. (2019) · Journal of Electromagnetic Waves and Applications · doi:10.1080/09205071.2019.1678525 · Cited by 4
“MultiScale Structural Similarity Index (MS-SSIM ) is an extended version of SSIM which supplies more flexibility than previous single-scale methods in incorporating the variations of viewing conditions [30].”
MethodJohari, M. M., & Behroozi, H. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9054432 · Cited by 4
“We incorporate MS-SSIM loss function to preserve luminance, contrast, and structural information of ground truth image in the produced image [23].”
MethodTexture Smoothing Quality Assessment via Information Entropy
Liu, C., Yang, C., Wei, M., & Wang, J. (2020) · IEEE Access · doi:10.1109/access.2020.2993146 · Cited by 6
End-to-End Deep ROI Image Compression
Akutsu, H., & Naruko, T. (2020) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2019edp7264 · Cited by 4
“In general, image quality measures such as PSNR (a mean squared error based metric) and MS-SSIM [8] (which qualifies structural similarities) are used as the loss function.”
MethodInfluentialA local structural information representation method for image quality assessment
Yang, X., Wang, T., & Ji, G. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09022-1 · Cited by 4
“And SSIM has many variants: multi-scale SSIM (MSSIM) [38], SSIM combined with the contrast sensitivity function (CSF + SSIM) [48].”
MethodEnd-to-end Optimized Video Compression with MV-Residual Prediction
Wu, X., Zhang, Z., Feng, J., et al. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00086 · Cited by 4
“In TUCODEC SSIM the loss D = 1 − Lmsssim is defined for the perceptional loss where Lmsssim is as defined in [16].”
BackgroundQuality Assessment for Retargeted Images: A Review
Karimi, M., & Entezami, E. (2020) · Iranian Conference on Machine Vision and Image Processing · doi:10.1109/mvip49855.2020.9116899 · Cited by 5
“Ordinary IQA methods [5,6,7] support just equal-sized images and they are not applicable for misaligned retargeted images [13].”
BackgroundMethodParallel Multi-threaded Gridrec Algorithm for Computer Tomography on GPU for Edge Computing
Chen, X., Zhu, Y., Zheng, X., et al. (2020) · IEEE International Conference on Cyber Security and Cloud Computing / IEEE International Conference on Edge Computing and Scalable Cloud (CSCloud/EdgeCom) · doi:10.1109/cscloud-edgecom49738.2020.00042 · Cited by 4
“Compared with PSNR or MSE, the structural similarity is more in line with the human eye's judgment of image quality in terms of image quality [17] .”
BackgroundKolganova, O., Tereshchenko, L., Сітко, А., et al. (2020) · International Conference on Advanced Computer Information Technologies (ACIT) · doi:10.1109/acit49673.2020.9208917 · Cited by 4
Pairwise Learning to Rank for Image Quality Assessment
Shi, Y., Niu, Y., Guo, W., Huang, Y., & Zhan, J. (2020) · IEEE Access · doi:10.1109/access.2020.3033122 · Cited by 6
“We compared the proposed PLRA index with 15 state-of-the-art FR-IQA metrics, including PSNR [1], PSNR-HVS [17], VSI [22], SSIM [2], IW-SSIM [67], MSSSIM [68], FSIM [21], RFSIM [69], SR-SIM [70], UQI [71], GSM [72], GSMD [25], VIF [24], IFC [73], MAD [15], and WaDIQaM-FR [32].”
MethodReducing latency and bandwidth for video streaming using keypoint extraction and digital puppetry
Prabhakar, R., Chandak, S., Chiu, C., Liang, R., Nguyen, H. M., Tatwawadi, K., & Weissman, T. (2021) · DCC · doi:10.1109/dcc50243.2021.00057 · Cited by 5
Data Adaptive Image Enhancement and Classification for Synthetic Aperture Sonar
Gerg, I. D., Williams, D. P., & Monga, V. (2020) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss39084.2020.9324047 · Cited by 4
“To achieve the goal of image enhancement, we use the multi-scale structural similar index measure (MS-SSIM) [12] as a prior for training our enhancement network.”
MethodOne-Shot GAN: Learning to Generate Samples from Single Images and Videos
Sushko, V., Gall, J., & Khoreva, A. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00293 · Cited by 7
“We observed that using λ = 0.15 leads to a good trade-off, re- DR SIFID ↓ LPIPS ↑ MS-SSIM ↓ Pixel ↑Lambda Diversity 0.00 0.05 0.04 0.95 0.33 0.05 0.07 0.19 0.77 0.50 0.15 0.08 0.33 0.63 0.66 0.50 0.13 0.39 0.55 0.69 Table C: Effect of the diversity regularization (DR) strength on DAVIS-YFCC100M in the Single Image set…”
InfluentialFootball Players Movement Analysis in Panning Videos
Działowski, K., & Forczmański, P. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-77977-1_15 · Cited by 4
“The image similarity was estimated using standard Structural SIM-ilarity (SSIM) method [29], assuming the perfect match is represented by value close to one.”
Han, Z., Li, L., Jin, W., Wang, X., Jiao, G., Liu, X., & Wang, H. (2021) · Sensors · doi:10.3390/s21113891 · Cited by 4
“Since the ground truth data could be obtained for the generated noisy data, fullreference image quality assessment (FRIQA) metrics were adopted to evaluate the denoising results of each algorithm, namely, the peak signal-to-noise-ratio (PSNR) [44] and structural similarity index measure (SSIM) [45].”
MethodEvaluating deep neural networks for image document enhancement
Kirsten, L. N., Piccoli, R., & Ribani, R. (2021) · ACM Symposium on Document Engineering (DocEng) · doi:10.1145/3469096.3474938 · Cited by 5
“The MSSSIM [27]2 is an IQA metric that evaluates the perceived change in”
BackgroundMethodInfluentialA No-Reference Image Quality Assessment Metric for Wood Images
Rajagopal, H., Mokhtar, N. B., Khairuddin, A. S. M., et al. (2021) · Journal of Robotics Networking and Artificial Life · doi:10.2991/jrnal.k.210713.012 · Cited by 4
“The proposed GGNR-IQA algorithm is compared with a commonly utilized NR-IQA, Blind/ Referenceless Image Spatial Quality Evaluator (BRISQUE) [18], and FR-IQAs namely, Structural Similarity Index (SSIM) [10], A RT I C L E I N F O Article History Received 19 October 2020 Accepted 01 June 2021 Keywords Wood images GLCM Ga…”
MethodInfluentialYu, X., Fan, J., Zhang, M., Liu, Q., Li, Y., Zhang, D., & Zhou, Y. (2021) · IEEE Access · doi:10.1109/access.2021.3101110 · Cited by 5
“The multiscale SSIM (MS-SSIM) method [45] applies SSIM evaluation at different scales.”
MethodA differentiable VMAF proxy as a loss function for video noise reduction
Ramsook, D., Kokaram, A. C., O'Connor, N. E., Birkbeck, N., Su, Y., & Adsumilli, B. (2021) · SPIE Applications of Digital Image Processing · doi:10.1117/12.2594164 · Cited by 4
“4,6,7 The Video Multimethod Assessment Fusion (VMAF) 8 is a complex Perceptual Video Quality Metric (PVQM) that uses image quality metrics such as Visual Information Fidelity (VIF), 9 Detail Loss Metric (DLM) 10 and temporal metrics such as Mean Co-Located Pixel Difference (MCPD).”
MethodYang, R., Liu, D., Ma, S., et al. (2021) · Proceedings - International Conference on Image Processing · doi:10.1109/icip42928.2021.9506765 · Cited by 4
“One of the most important advantages of end-to-end image compression is that it can easily improve perceptual quality via changing the loss function, such as multi-scale structural similarity (MS-SSIM) [8].”
BackgroundReduction of JPEG Artifacts using BSDEs
Borkowski, D. (2021) · Computer Science Research Notes · doi:10.24132/csrn.2021.3101.1 · Cited by 4
Chin, D. J. Y., Mohamed, A. S. A., Shariff, K. A., et al. (2021) · Sensors · doi:10.3390/s21237955 · Cited by 4
“Both metrics measure the structural similarity over the 2D images, with the main difference in MS-SSIM being that it measures over different image scales through different image sub-sampling processes with low-pass filters [28].”
Application of the XBoost Regressor for an A Priori Prediction of UAV Image Quality
Sekrecka, A. (2021) · Remote Sensing · doi:10.3390/rs13234757 · Cited by 4
Subjective and Objective Quality Assessment of Swimming Pool Images
Lei, F., Li, S., Xie, S., & Liu, J. (2022) · Frontiers in Neuroscience · doi:10.3389/fnins.2021.766762 · Cited by 4
“• MS-SSIM pointed out by Wang et al. (2003), performs the SSIM in different scales and integrates their outputs with psychophysical weights.”
MethodAttention Mechanism-Based Light-Field View Synthesis
Gul, M. S. K., Mukati, M. U., Bätz, M., et al. (2022) · IEEE Access · doi:10.1109/access.2022.3142949 · Cited by 4
“Best results are highlighted. multi-scale structural similarity index (MS-SSIM) [66] also in dB according to the following equation : The objective evaluation is conducted directly on the RGB images as well as on the luminance channel only.”
MethodColor Pixel Reconstruction for a Monolithic RGB-Z CMOS Imager
Rebiere, V., Drouot, A., Granado, B., et al. (2022) · Journal of Signal Processing Systems · doi:10.1007/s11265-021-01726-3 · Cited by 4
Lee, W., Cho, E., Kim, W., et al. (2022) · SPIE Medical Imaging: Image Perception, Observer Performance, and Technology Assessment · doi:10.1117/12.2612541 · Cited by 4
“[5], which includes root mean square error (RMSE), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), multi-scale (MS)-SSIM [9], information content-weighted (IW)-SSIM [10], gradient magnitude similarity deviation (GMSD [11]), feature similarity index (FSIM [12]), noise quality measure (NQM, [13]),…”
BackgroundHe, D., Zhong, Y., Shi, Q., Liu, X., & Zhang, L. (2022) · Science of Remote Sensing · doi:10.1016/j.srs.2022.100041 · Cited by 4
A multi-level feature integration network for image inpainting
Chen, T., Zhang, X., Hamann, B., et al. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-13028-2 · Cited by 4
“In order to increase structural similarity between inpainted images and the ground truth, we incorporate multi-scale structural similarity (MS SSIM) loss [35].”
MethodSemi-Supervised Pseudo-Healthy Image Synthesis via Confidence Augmentation
Du, Y., Quan, Q., Han, H., et al. (2022) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi52829.2022.9761522 · Cited by 4
“The metrics are defined as follows: ) where S pred is a pre-trained lesion predictor, N counts the ”pathological” ones predicted by S pred and MS − SSIM stands for multiscale structural similarity [11].”
BackgroundMantiuk, R., Dénes, G., Chapiro, A., et al. (2021) · ACM Transactions on Graphics · doi:10.1145/3476576.3476595 · Cited by 4
“Most popular image quality metrics, such as PSNR and MS-SSIM [Wang et al. 2003], take as input a pair of gamma-encoded images and ignore all aspects of their physical presentation, including screen resolution, size, viewing distance, and display peak luminance.”
BackgroundVMQ: an algorithm for measuring the Video Motion Quality
Alsrehin, N. O., & Klaib, A. F. (2019) · Bulletin of Electrical Engineering and Informatics · doi:10.11591/eei.v8i1.1418 · Cited by 4
Saini, L. K., & Mathur, P. (2022) · Journal of medical imaging · doi:10.1117/1.jmi.9.5.052403 · Cited by 7
A Diverse Domain Generative Adversarial Network for Style Transfer on Face Photographs.
Tahir, R., Cheng, K., Memon, B. A., & Liu, Q. (2022) · International Journal of Interactive Multimedia and Artificial Intelligence · doi:10.9781/ijimai.2022.08.001 · Cited by 4
A Domain-Guided Model for Facial Cartoonlization
Yang, N., Xia, B., Han, Z. J., et al. (2022) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2022.105887 · Cited by 4
Development of a landslide detection surveillance system
Agrawal, A., & Kar, S. (2022) · International Conference on Systems, Signals and Image Processing · doi:10.1109/iwssip55020.2022.9854417 · Cited by 4
Clutter Removal in Ground-Penetrating Radar Images Using Deep Neural Networks
Sun, H.-H., Cheng, W., & Fan, Z. (2022) · doi:10.1109/isap53582.2022.9998650 · Cited by 4
A Low Latency and Compact GCD Design Using an Intelligent Seed-Selection Scheme of LL-PRNG
Mitra, S., Goenka, A. G., & Naskar, M. K. (2023) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2023.3251746 · Cited by 6
“9(e) the average multi-SSIM [16] of the retrieved images is calculated as 98.86% and Fig.”
MethodInfluentialStochastic surprisal: An inferential measurement of free energy in neural networks
Prabhushankar, M., & AlRegib, G. (2023) · Frontiers in Neuroscience · doi:10.3389/fnins.2023.926418 · Cited by 4
“, 2004), MS-SSIM (Wang et al., 2003), CWSSIM (Sampat et al.”
Wu, R., Liu, G., Lv, J., et al. (2023) · Remote Sensing · doi:10.3390/rs15082119 · Cited by 4
Le, M., & Carlsson, N. F. (2023) · ACM Conference on Data and Application Security and Privacy (CODASPY) · doi:10.1145/3577923.3583645 · Cited by 6
“MS-SSIM is a variation of SSIM, where the measurements are conducted by using multiple scales of sub-sampling.”
BackgroundMethodInfluentialA Novel Image Similarity Measure Based on Greatest and Smallest Eigen Fuzzy Sets
Martino, F. D., & Sessa, S. (2023) · Symmetry · doi:10.3390/sym15051104 · Cited by 4
Image Quality Assessment for Realistic Zoom Photos
Han, Z., Liu, Y., Xie, R., & Zhai, G. (2023) · Sensors · doi:10.3390/s23104724 · Cited by 7
“As research in quality assessment has demonstrated that incorporating multi-scale information correlates better with human perception [69,70], we extract the above features in two scales (low-pass filtered and downsampled by 2).”
BackgroundVisual Quality Assessment of Point Clouds Compared to Natural Reference Images
Baek, A., Kim, M., Son, S., et al. (2023) · Journal of Web Engineering · doi:10.13052/jwe1540-9589.2232 · Cited by 4
Learned Image Compression with Wavelet Preprocessing for Low Bit Rates
Iliopoulou, S., Tsinganos, P., Ampeliotis, D., & Skodras, A. N. (2023) · IEEE International Conference on Digital Signal Processing (DSP) · doi:10.1109/dsp58604.2023.10167974 · Cited by 4
“The distortion is the expected difference between the reconstructed and the original image, as measured by a norm or perceptual metric function, e.g., the mean squared error (MSE) or the multi-scale structural similarity (MS-SSIM) [20].”
MethodCCGAN as a Tool for Satellite-Derived Chlorophyll a Concentration Gap Reconstruction
Ćatipović, L., Matić, F., Kalinić, H., et al. (2023) · Journal of Marine Science and Engineering · doi:10.3390/jmse11091814 · Cited by 4
“…was proposed, whose generator loss function has been swapped for: The 2 function has been replaced with Structural Similarity Index function ( ψ ) [39,40]. ψ is defined where a and b are square matrices, µ a and µ b are mean values of a and b , σ 2 a and σ 2 b are variances of a and b , σ ab is…”
BackgroundInfluentialEHFP-GAN: Edge-Enhanced Hierarchical Feature Pyramid Network for Damaged QR Code Reconstruction
Zheng, J., Zhao, R., Lin, Z., Liu, S., Zhu, R., Zhang, Z., Fu, Y., & Lu, J. (2023) · Mathematics · doi:10.3390/math11204349 · Cited by 7
“Structural similarity loss Loss SSIM [34] is calculated using the following formula: where µ x and µ y represent the mean values of the original and generated images, respectively, σ 2 x and σ 2 y represent the variances of the original and generated images, respectively, σ xy represents the…”
MethodResultJPEG Image Enhancement with Pre-Processing of Color Reduction and Smoothing
Shoda, A., Miyazaki, T., & Omachi, S. (2023) · Sensors · doi:10.3390/s23218861 · Cited by 4
Assessing quality assurance of multi‐leaf collimator using the structural similarity index
Zhang, H., Zhang, B., Lasio, G., et al. (2024) · Journal of Applied Clinical Medical Physics · doi:10.1002/acm2.14288 · Cited by 4
AML leukocyte classification method for small samples based on ACGAN
Zhang, C., & Zhu, J. (2024) · Biomedizinische Technik/Biomedical Engineering · doi:10.1515/bmt-2024-0028 · Cited by 4
StyleAdv: A Usable Privacy Framework Against Facial Recognition with Adversarial Image Editing
Le, M., & Carlsson, N. F. (2024) · Proceedings on Privacy Enhancing Technologies · doi:10.56553/popets-2024-0043 · Cited by 6
A Generative Adversarial Network AMS-CycleGAN for Multi-Style Image Transformation
Rang, X., Dong, Z., Han, J. J., et al. (2024) · IEEE Access · doi:10.1109/access.2024.3397492 · Cited by 4
“To capture the details and structural information of images and enhance the similarity of reconstructed images, the MS-SSIM [33] loss function is introduced into the reconstruction loss.”
BackgroundMethodSLIC: A Learned Image Codec Using Structure and Color
Prativadibhayankaram, S., Panda, M. P., Richter, T., et al. (2024) · DCC · doi:10.1109/dcc58796.2024.00008 · Cited by 4
“Loss Function: As distortion metrics, we use mean squared error (MSE) and multi-scale structural similarity index measure (MS-SSIM) [14] for structural fidelity.”
MethodANNLR-FSR: Artificial Neural Network-Based Locality Regularization for Face Super-Resolution
Rai, D., & Rajput, S. S. (2024) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2024.3408832 · Cited by 4
Neural activity shaping in visual prostheses with deep learning
Castro, D., Grayden, D. B., Hamish Meffin, A., & Spencer, M. (2024) · Journal of Neural Engineering · doi:10.1088/1741-2552/ad6186 · Cited by 6
“Here, we used a standard MSE, but we speculate that metrics sensitive to the information content of the image (such as Structural Image Index (Wang et al 2003)) may lead to better outcomes.”
Benchmarking Robustness in Neural Radiance Fields
Wang, C., Wang, A., Li, J., Yuille, A., & Xie, C. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00298 · Cited by 11
“RmCM for PSNR, SSIM and LPIPS can be found from Table 9 to Table 14.”
MethodInfluentialSingh, P. K., & Chandler, D. M. (2013) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2005153 · Cited by 4
Study on Signal-to-noise ratio algorithms based on no-reference Image quality assessment
Zhu, X., Li, X., Li, Z., et al. (2012) · International Conference on Systems and Informatics (ICSAI) · doi:10.1109/icsai.2012.6223383 · Cited by 4
Full Reference Image Quality Assessment: Limitation
Chetouani, A. (2014) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2014.153 · Cited by 4
Data-Driven Modules for Objective Visual Quality Assessment Focusing on Benchmarking and SLAs
Barri, A., & Dooms, A. (2016) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2016.2637164 · Cited by 4
Khosravi, M. H., Hassanpour, H., & Ahmadifard, A. (2017) · Multimedia Tools and Applications · doi:10.1007/s11042-017-4636-7 · Cited by 5
“From the literature, it is well-known that the recognizability of image contents depends on its resolution, the distance from the image screen to the observer, and the perceptual capability of the observer’s visual system [47].”
BackgroundMethodInformation fusion based techniques for HEVC
Fernández, D. G., Barrio, A. A. D., Botella, G., et al. (2017) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2262604 · Cited by 4
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.