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 2,901–3,000 of 6,047
A Psychophysical Evaluation of Texture Compression Masking Effects
Lavoué, G., Langer, M., Peytavie, A., et al. (2018) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2018.2805355 · Cited by 7
HEVC encoder optimization for HDR video coding based on irregularity concealment effect
Yu, S., Jung, C., & Lin, Q. (2018) · Signal Processing Image Communication · doi:10.1016/j.image.2018.02.008 · Cited by 8
Golestaneh, S. A., & Karam, L. J. (2018) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2018.00117 · Cited by 7
“Natural image statistics and structural similarity are used in existing popular objective image quality assessment (IQA) methods that are designed for natural images [40, 43, 42, 25, 24, 48].”
BackgroundMethodAmor, M. B., Kammoun, F., & Masmoudi, N. (2018) · Computer Standards & Interfaces · doi:10.1016/j.csi.2018.04.003 · Cited by 7
On the possibility to achieve 6-DoF for 360 video using divergent multi- view content
Ray, B., Jung, J., & Larabi, M. (2018) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco.2018.8553397 · Cited by 7
“For objective quality evaluation, PSNR and MS-SSIM metric are used [11].”
MethodMethodology for fine-grained monitoring of the quality perceived by users on 360VR contents
Muñoz, L., Díaz, C., Orduna, M., Ronda, J. I., Pérez, P., Benito, I., & Garcia, N. N. (2020) · Digital Signal Processing · doi:10.1016/j.dsp.2020.102706 · Cited by 7
No-reference image quality assessment based on AdaBoost BP neural network in wavelet domain
Yan, J., Bai, X., Zhang, W., Xiao, Y., Chatwin, C. R., Young, R. C. D., & Birch, P. M. (2019) · Journal of Systems Engineering and Electronics · doi:10.21629/jsee.2019.02.01 · Cited by 7
“High quality natural images have regular statistical characteristics, and the distortions can alter the image structure as well as the statistical characteristics [23,24].”
BackgroundPractical Use Suggests a Re-evaluation of HDR Objective Quality Metrics
Sugito, Y., & Bertalmı́o, M. (2019) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex.2019.8743300 · Cited by 9
Accelerating Redundant DCT Filtering for Deblurring and Denoising
Fukushima, N., Kawasaki, Y., & Maeda, Y. (2019) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2019.8803511 · Cited by 7
Learning to Evaluate Color Similarity for Histopathology Images using Triplet Networks
Choudhary, A., Wu, H., Li, T., & Wang, M. D. (2019) · ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB) · doi:10.1145/3307339.3342170 · Cited by 7
“…(HSV/ Lαβ ) or leveraging perceptually motivated distance metrics used for image quality assessment like Structural Similarity Index (SSIM) [10], Feature Similarity Index (FSIM) [53], Multi-Scale Structural Similarity Index (MS-SSIM) [47] and Quante-rion Structural Similarity Index (QSSIM) [25].”
MethodFull-reference IPTV image quality assessment by deeply learning structural cues
Kong, Y., Liu, C., & Hou, R. (2020) · Signal Processing Image Communication · doi:10.1016/j.image.2020.115779 · Cited by 7
Full Reference Video Quality Measures Improvement Using Neural Networks
Tiotsop, L. F., Servetti, A., & Masala, E. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053739 · Cited by 7
“Mapping features to perceived visual quality has been done in different ways, for instance using a weighted combination of the features [7, 8], through a ratio between information extracted from the PVS and the SRC to measure how distorted the PVS is [9] or exploiting machine learning algorithms.”
MethodHigh-Resolution Crowd Density Maps Generation With Multi-Scale Fusion Conditional GAN
Huang, S., Zhou, H., Liu, Y., & Chen, R. (2020) · IEEE Access · doi:10.1109/access.2020.3000741 · Cited by 6
“The values of exponents α m , β i , γ i are the same as those in [45].”
MethodResultLaaroussi, S., Baataoui, A., Halli, A., et al. (2020) · IET Image Processing · doi:10.1049/iet-ipr.2019.1619 · Cited by 7
“To summarise the results in Table 1, the proposed approach obtained the highest values for MS-SSIM and SSIM-seamline while being either the first or second for the SSIM measure.”
MethodInfluentialVandecappelle, M., Vervliet, N., & Lathauwer, L. D. (2020) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2020.3045911 · Cited by 10
“In Table V, the peak SNR (PSNR) and structural similarity index (SSI) [72] [73].”
BackgroundMulti-Feature 360 Video Quality Estimation
Azevedo, R. G. D. A., Birkbeck, N., Janatra, I., Adsumilli, B., & Frossard, P. (2021) · IEEE Open Journal of Circuits and Systems · doi:10.1109/ojcas.2021.3073891 · Cited by 10
“For that, we computed the following objective metrics in both the projection and viewport domains: PSNR, PSNR-HVS, PSNR-HVSM, SSIM, MSSSIM, GMSD, Spatial activity, and Temporal activity.”
BackgroundInfluentialA Boundary-aware Distillation Network for Compressed Video Semantic Segmentation
Lu, H., & Deng, Z. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412821 · Cited by 7
“where k denotes the k stage, lBCE is the BCE loss, IIOU indicates the IOU loss [22] and lSSIM is the SSIM loss [23].”
BackgroundOverparametrization of HyperNetworks at Fixed FLOP-Count Enables Fast Neural Image Enhancement
Müller, L. K. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00038 · Cited by 6
SR-Inpaint: A General Deep Learning Framework for High Resolution Image Inpainting
Xu, H., Li, X., Zhang, K., et al. (2021) · Algorithms · doi:10.3390/a14080236 · Cited by 7
“Among them, PSNR is used to measure the degree of deformation and noise; SSIM is used to describe the degree of similarity of the graphics structure; NRMSE is used to measure the pixel error; FID is used to measure the perceptual error based on deep features, using a pre-trained Inception-V3 model [50].”
MethodAMDFNet: Adaptive multi-level deformable fusion network for RGB-D saliency detection
Li, F., Zheng, J., Zhang, Y., Liu, N., & Jia, W. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.08.116 · Cited by 7
ERQA: Edge-restoration Quality Assessment for Video Super-Resolution
Kirillova, A., Lyapustin, E., Antsiferova, A., & Vatolin, D. S. (2022) · International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP) · doi:10.5220/0010780900003124 · Cited by 18
“We calculated several well-known metrics on a new dataset: PSNR, SSIM (Wang et al., 2004), MS-SSIM (Wang et al., 2003), VMAF6, the recently developed LPIPS (Zhang et al., 2018), which showed good results when assessing super-resolution imaging, its improvement DISTS (Ding et al., 2020b) and metric for SR assessment (M…”
MethodOn the Link Between Subjective Score Prediction and Disagreement of Video Quality Metrics
Tiotsop, L. F., Agboma, F., Wallendael, G. V., Aldahdooh, A., Bosse, S., Janowski, L., Barkowsky, M., & Masala, E. (2021) · IEEE Access · doi:10.1109/access.2021.3127395 · Cited by 7
Lam, N. F. D., Sun, H., Song, L., Yang, D., Zhi, S., Ren, G., Chou, P.-H., Wan, S. B. N., Wong, M. F. E., Chan, K. K., et al. (2022) · Quantitative Imaging in Medicine and Surgery · doi:10.21037/qims-21-791 · Cited by 6
A Regression-Based Family of Measures for Full-Reference Image Quality Assessment
Oszust, M. (2016) · Measurement Science Review · doi:10.1515/msr-2016-0040 · Cited by 7
“Similar observations were made for rSIM2 (FSIM, FSIMc, GSM, MSSIM, SR-SIM, and SFF), rSIM3 (FSIM, FSIMc, IW-SSIM, PSNR, SSIM), and rSIM4 (VSI, FSIM, FSIMc, MSSIM, RFSIM, SR-SIM, SSIM, and SFF).”
BackgroundMethodInfluentialAnalysis of reported error in Monte Carlo rendered images
Whittle, J., Jones, M. W., & Mantiuk, R. (2017) · The Visual Computer · doi:10.1007/s00371-017-1384-7 · Cited by 11
“These models vary from a simple linear weighting of features in a multi-component error measure [52,54] to models based on a nonlinear contrast sensitivity function [33] applied at multiple scales in an MGA decomposition.”
MethodAn Error-Activation-Guided Blind Metric for Stitched Panoramic Image Quality Assessment
Yang, L., Liu, J., & Gao, C. (2017) · Communications in computer and information science · doi:10.1007/978-981-10-7302-1_22 · Cited by 7
“Because of the uniqueness of two most common error types during image stitching—shape breakage (including object fragment) and ghosting, stitched image quality assessment is fundamentally different from conventional IQA for images distorted from compression artifacts or network packet losses [17,18,24].”
BackgroundPerformance Comparison of High Efficiency Video Coding (HEVC) with H.264 AVC
Ram, C., & Panwar, S. (2017) · International Conference on Signal-Image Technology & Internet-Based Systems (SITIS) · doi:10.1109/sitis.2017.58 · Cited by 7
“The structural similarity video quality paradigm is build on the assumption that the human visual system is highly habituated for extracting structural information from the video scene, and therefore an estimation of structural similarity index can provide a valuable approximation to perceived video quality (see [44]…”
BackgroundAn image quality assessment algorithm based on saliency and sparsity
Banitalebi-Dehkordi, M., Khademi, M., Ebrahimi-Moghadam, A., & Hadizadeh, H. (2018) · Multimedia Tools and Applications · doi:10.1007/s11042-018-6700-3 · Cited by 10
“We compare the results against the state-of-the-art objective metrics such as UQI [21], NQM [19], IFC [46], VSNR [17], MS-SSIM [51], VSM [57] and WASH [44].”
MethodResultInfluentialIterative Adaptive Nonconvex Low-Rank Tensor Approximation to Image Restoration Based on ADMM
Shen, Z., & Sun, H. (2018) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-018-0867-0 · Cited by 7
“The SSIM index is a measure for predicting the experiment’s quality and representing the similarity of images, more details about SSIM discussed in [51].”
MethodAbazari, R., & Lakestani, M. (2019) · Informatica · doi:10.15388/informatica.2019.194 · Cited by 7
Chung, D., & Delp, E. J. (2019) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2019.00193 · Cited by 7
“We defined MS-SSIM [35] as Full SP-StarGAN loss function.”
MethodGenerative collaborative networks for single image super-resolution
Seddik, M. E. A., Tamaazousti, M., & Lin, J. Y. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.02.068 · Cited by 9
“In fact, since MSE consists in a pixel-wise images differences, its ability to recover high texture details is limited [1, 5, 6, 7].”
BackgroundImproving Discrete Latent Representations With Differentiable Approximation Bridges
Ramapuram, J., & Webb, R. (2020) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn48605.2020.9207581 · Cited by 8
“objectives through the use of two metrics: the MS-SSIM [54] and linear classification of posterior samples.”
MethodInfluentialNon‐uniform image blind deblurring by two‐stage fully convolution network
Wu, C., Wo, Y., Han, G., Wu, Z., & Liang, J. (2020) · IET Image Processing · doi:10.1049/iet-ipr.2018.5716 · Cited by 6
“We fed all the testing samples into proposed PG-net to output the restored images, and evaluate it with six commonly used image quality metrics suggested by [30]: Peak Signal to Noise Ratio (PSNR), Structural SIMilarity (SSIM) [35], Multi-Scale Structural SIMilarity (MS-SSIM) [36], Information Fidelity Criterion (IFC)…”
MethodA GAN-Based Video Intra Coding
Zhong, G., Wang, J., Hu, J., et al. (2021) · Electronics · doi:10.3390/electronics10020132 · Cited by 7
“In [30], a machine learning-based video codec in low-latency mode is presented and surpasses all commercial video codecs in terms of Multi-Scale Structural Similarity Index (MS-SSIM) [32].”
MethodSpatial Constant Quantization in JPEG XR is Nearly Optimal
Richter, T. (2010) · DCC · doi:10.1109/dcc.2010.14 · Cited by 7
“While independent works [10] already reported a considerable improvement in terms of the SSIM [3], [4] image quality index, it will be seen here that no such improvements should be expected in terms of PSNR for algorithms of feasible complexity.”
BackgroundPerformance Evaluation of Image Retrieval System Based on Error Metrics
Jeyakumar, V., & Kanagaraj, B. R. (2015) · Indian Journal of Science and Technology · doi:10.17485/ijst/2015/v8is7/64950 · Cited by 8
DSSLIC: Deep Semantic Segmentation-based Layered Image CompressionPreprint
Akbari, M. A., Liang, J., & Han, J. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.03348 · Cited by 7
“It is shown that when the bit rate is less than about 1.4 bpp, our scheme achieves better results than other codes in both PSNR and MS-SSIM.”
MethodInfluentialConvolutional Neural Networks for Video Quality AssessmentPreprint
Giannopoulos, M., Tsagkatakis, G., Blasi, S., Toutounchi, F., Mouchtaris, A., Tsakalides, P., Mrak, M., & Ebroul Izquierdo, M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1809.10117 · Cited by 9
“When dealing with variations across scales, SSIM can be extended to Multi-Scale Structural Similarity (MS-SSIM) [8].”
MethodGated Context Model with Embedded Priors for Deep Image CompressionPreprint
Liu, H., Chen, T., Guo, P., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1902.10480 · Cited by 7
“We replaced the MSE with MS-SSIM [28] because it exhibits better correlation with the subjective quality perceived by our HVS.”
MethodJoint Frequency- and Image-Space Learning for Fourier Imaging.Preprint
Singh, N., Iglesias, J. E., Elfar Adalsteinsson, A., Dalca, A. V., & Golland, P. (2020) · arXiv (Cornell University) · Cited by 9
Deep Neural Networks for Blind Image Quality Assessment: Addressing the Data ChallengePreprint
Athar, S., Wang, Z., & Wang, Z. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.12161 · Cited by 8
“The BIQA method dipIQ [31] uses three FR methods (GMSD [100], MSSSIM [102], and VIF [101]) to generate 80-million quality-discriminable image pairs (DIPs) from a dataset that has 840 reference and 16,800 distorted images, which are then used to learn a blind quality model.”
MethodSelf-citationData Augmentation For Medical MR Image Using Generative Adversarial NetworksPreprint
Huang, P., Liu, X., & Huang, Y. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.14297 · Cited by 15
“• Brain Tumor MRI Generation: This research shows that PGGAN-based with new loss function is able to generate promising results, exceeding existing GAN-based meth-ods both in FID [18] and MS-SSIM [19].”
ResultConditional Diffusion Models for Semantic 3D Medical Image SynthesisPreprint
Dorjsembe, Z., Pao, H.-K., Odonchimed, S., & Xiao, F. (2023) · doi:10.36227/techrxiv.23723787 · Cited by 10
Synthetic Infra-Red Image Evaluation Methods by Structural Similarity Index Measures
Lee, S. H., & Leeghim, H. (2022) · Electronics · doi:10.3390/electronics11203360 · Cited by 6
Encoder-Quantization-Motion-based Video Quality Metrics
Chen, Y., Shang, Z., Wei, H., et al. (2024) · Picture Coding Symposium · doi:10.1109/pcs60826.2024.10566470 · Cited by 6
Zhou, Y., Chen, G., & Xu, C. (2024) · Optics and Lasers in Engineering · doi:10.1016/j.optlaseng.2024.108549 · Cited by 6
Ribeiro, R. D. L., Carvalho, A., & Rodrigues, N. (2024) · IEEE Transactions on Games · doi:10.1109/tg.2024.3487054 · Cited by 13
“Its modern variation, multiscale structural similarity indexing measure (MS-SSIM) [115], performs the evaluation at multiple scales through a multistep downsampling process.”
WGAN-Based Realization Process of Gravel Soil for Hydraulic Property Simulation
Bin, Z., & Hu, X.-G. (2024) · Applied Sciences · doi:10.3390/app14219873 · Cited by 6
High-Resolution SAR-to-Multispectral Image Translation Based on S2MS-GAN
Liu, Y., Han, Q., Hong, Y., & Hu, H. (2024) · Remote Sensing · doi:10.3390/rs16214045 · Cited by 8
Ou, Z., Pan, Y., Xie, F., Guo, Q., & Shen, D. G. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3492020 · Cited by 8
Unrolled deep learning for breast cancer detection using limited-view photoacoustic tomography data
John, M., & Barhumi, I. (2025) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-025-03302-4 · Cited by 6
BackdoorBench: A Comprehensive Benchmark and Analysis of Backdoor Learning
Wu, B., Chen, H., Zhang, M., Zhu, Z., Wei, S., Yuan, D., Zhu, M., Wang, R., Liu, L., & Shen, C. (2025) · International Journal of Computer Vision · doi:10.1007/s11263-025-02447-x · Cited by 17
“We believe that these analyses could shed light on the backdoor learning mechanism, to facilitate us to develop more effective backdoor learning algorithms or build backdoor related theories in future.”
Coarse-to-Fine: A Dual-Phase Channel-Adaptive Method for Wireless Image Transmission
Li, H., Zhang, G., Zhou, K., Cai, Y., & Yu, G. (2025) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2025.3597717 · Cited by 9
Cone Beam Computed Laminography Based on Generalized Charbonnier Penalty
Liu, Y., Tang, H., Lu, J., Zhang, P., Gui, Z., Li, S., Guo, S., & Liu, Y. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3617418 · Cited by 8
“The root-mean-square error (RMSE) [44], multiscale structural similarity (MSSIM) [45], Pearson correlation coef-ficient (PCC) [46], and universal image quality index (UQI) [47] of the reconstructed images were calculated and compared to objectively evaluate the reconstruction accuracy of di ff erent…”
Leveraging Envelope Data in cGAN for Robust GPR Inversion
Huang, M., Wang, Y., Wu, Y., et al. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3631324 · Cited by 6
“Furthermore, we propose a combination of MSE and MS-SSIM [43] as a hybrid loss function.”
An Azimuth-Continuously Controllable SAR Image Generation Algorithm Based on GAN
Cui, Y., Liu, Z., Ruan, L., et al. (2025) · Remote Sensing · doi:10.3390/rs17223763 · Cited by 6
“In this study, the parameters for MS-SSIM are configured based on the work of Wang et al. [35].”
Zhou, Y., Yang, Q., Ren, K., & Chen, Q. (2025) · Infrared Physics & Technology · doi:10.1016/j.infrared.2025.106325 · Cited by 10
AIM 2025 Low-Light RAW Video Denoising Challenge: Dataset, Methods and Results
Yakovenko, A., Chakvetadze, G., Khrapov, I., Zhelezov, M. S., Vatolin, D., Timofte, R., Oh, Y., Kwon, J.-H., Park, J., Cho, N. I., et al. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00596 · Cited by 20
A full-reference stereoscopic image quality metric based on binocular energy and regression analysis
Galkandage, C., Calic, J., Silva, D., et al. (2015) · 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video · doi:10.1109/3dtv.2015.7169375 · Cited by 6
“Well established metrics such as Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index (SSIM) [1] have been used to build algorithms for stereoscopic quality assessment.”
MethodRetina model inspired image quality assessment
Zhai, G., Kaup, A., Wang, J., et al. (2013) · Visual Communications and Image Processing · doi:10.1109/vcip.2013.6706367 · Cited by 6
“MS-SSIM 0.”
MethodInfluentialA database of local masking thresholds in natural images
Alam, M. M., Vilankar, K. P., & Chandler, D. M. (2013) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2008581 · Cited by 6
A Perceptually Lossless Image Compression Scheme Based on JND Refinement by Neural Network
Lie, W., & Liu, W.-C. (2010) · Pacific-Rim Symposium on Image and Video Technology · doi:10.1109/psivt.2010.44 · Cited by 6
“For examples, Memon and Moayeri [7] and other works [8][9] proposed new definitions of nearlossless (different from JPEG-LS) in order to better match with the human perceptual system.”
BackgroundBlind Stereopair Quality Assessment Using Statistics of Monocular and Binocular Image Structures
Fan, Y., Larabi, M., & Cheikh, F. A. (2019) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2019.8802956 · Cited by 7
“Furthermore, several previous works showed that the image scale affects quality prediction [29, 30].”
BackgroundMethodWeakly-Supervised Photo-realistic Texture Generation for 3D Face Reconstruction
Yin, X., Huang, D., Fu, Z., Wang, Y., & Chen, L. (2023) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg57933.2023.10042545 · Cited by 9
“For the reconstruction ability evaluation, we calculate the L1 loss and the structural similarity (Wang, Simoncelli, and Bovik 2003) of the reconstructed face images.”
MethodCNN-based morphological decomposition of X-ray images for details and defects contrast enhancement
Madmad, T., Delinte, N., & Vleeschouwer, C. D. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00246 · Cited by 7
“We adopt the Multi-scale Structural Similarity Index Measure (MS-SSIM) [53] for each loss.”
MethodSemantic-aware deidentification generative adversarial networks for identity anonymization
Kim, H., Pang, Z., Zhao, L., Su, X., & Lee, J. S. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-13917-6 · Cited by 7
“In addition, a structural similarity index (SSIM) loss [45] and a perceptual loss [17] are added to the objective function to improve the quality of the generated image.”
BackgroundMethodRegional realness-aware generative adversarial networks for stain normalization
Kablan, E. B. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-08659-x · Cited by 6
“The performance of the proposed method was measured with different evaluation metrics including: Structural Similarity Index (SSIM) [47], Feature Similarity (FSIM) Index [48], Peak Signal-to-Noise Ratio (PSNR), Mean Squared Error (MSE), Root Mean Square Error (RMSE) [49], Multi-scale Structural Similarity Index (MSSSI…”
MethodThe disparity between optimal and practical Lagrangian multiplier estimation in video encoders
Ringis, D. J., Vibhoothi, V., Pitié, F., & Kokaram, A. C. (2023) · Frontiers in Signal Processing · doi:10.3389/frsip.2023.1205104 · Cited by 6
A Super-Resolution Algorithm Based on Hybrid Network for Multi-Channel Remote Sensing Images
Li, Z., Zhang, W., Pan, J., et al. (2023) · Remote Sensing · doi:10.3390/rs15143693 · Cited by 6
Adversarial image generation by spatial transformation in perceptual colorspaces
Aydın, A., & Temizel, A. (2023) · Pattern Recognition Letters · doi:10.1016/j.patrec.2023.09.003 · Cited by 7
“…by exploring YC b C r space as well as perceptually uniform CIELAB space and measuring structured similarity metrics such as Structured Similarity Index Measure (SSIM) [22] and Multi-Scale Structured Similarity Index Measure (MS-SSIM) [23] between benign images and adversarially generated images.”
BackgroundMethodZhang, Z., Ao, Z., Wu, W., Wang, Y., & Xin, Q. (2024) · Remote Sensing · doi:10.3390/rs16061086 · Cited by 6
Synchrotron radiation sparse-view CT artifact correction through deep learning neural networks
Huang, M., Li, G., Zhang, J., et al. (2024) · Nondestructive Testing And Evaluation · doi:10.1080/10589759.2024.2334431 · Cited by 6
“The Peak signal-to-noise ratio(PSNR) and the multi-scale structural similarity MS-SSIM [26] were used to assessing the predicted images quantitatively.”
Self-citationFactorized Motion Fields for Fast Sparse Input Dynamic View Synthesis
Somraj, N., Choudhary, K., Mupparaju, S. H., et al. (2024) · ACM SIGGRAPH Conference Papers (SIGGRAPH) · doi:10.1145/3641519.3657498 · Cited by 6
“Evaluation measures : We evaluate the rendered frames of all the methods using PSNR, SSIM [Wang et al. 2003] and LPIPS [Zhang et al. 2018].”
MethodShu, R., & Zhao, S. (2024) · Sensors · doi:10.3390/s24144522 · Cited by 8
“The Structure Similarity Index Measure (SSIM) [19] implements constraints through the similarity of the structural information of images, which, in addition to its important role in assessing image quality [20], also shows potential in optimizing saliency map prediction [21], rate control [22],…”
BackgroundAdaptive Weighting via Federated Evaluation Mechanism for Domain Adaptation with Edge Devices
Zhao, R., Yang, X. W., Zhi, P., Zhou, R., Zhou, Q., & Jin, Q. (2024) · ACM Transactions on Sensor Networks · doi:10.1145/3669903 · Cited by 8
ConIQA: A deep learning method for perceptual image quality assessment with limited data
Eybposh, M. H., Cai, C., Moossavi, A., Rodríguez-Romaguera, J., & Pégard, N. C. (2024) · Scientific Reports · doi:10.1038/s41598-024-70469-5 · Cited by 12
“We considered popular and high-performance FR-IQA metrics: LPIPS 38 , Mean Square Error (MSE) 31 , Mean Absolute Error (MAE) 31 , Root Mean Squared Error (RMSE) 31 , Peak Signal-to-Noise Ratio (PSNR) 31 , Cosine similarity (Cosine) 28 , Structural Similarity Index Measure (SSIM) 32,54 , Multi-Scale SSIM (MS-SSIM) 55 ,…”
BackgroundOverview of High-Dynamic-Range Image Quality Assessment
Liu, Y., Tian, Y., Wang, S., Zhang, X., & Kwong, S. (2024) · Journal of Imaging · doi:10.3390/jimaging10100243 · Cited by 8
“…F-test, as suggested in [76], across five publicly available databases, to evaluate the performance of three perceptual mapping methods (Gamma, PQ, PU21) combined with nine LDR IQA metrics (PSNR, SSIM [27], MS-SSIM [77], FSIM [78], GMSD [76], VIF [79], ESIM [80], GFM [81], and DISTS [43]).”
ROI-DVC: A Region-of-Interest Based Deep Video Coding Framework
Wu, X., Wang, P., & Wang, X. (2024) · Proceedings - International Conference on Image Processing · doi:10.1109/icip51287.2024.10647832 · Cited by 12
IterInv: Iterative Inversion for Pixel-Level T2I Models
Tang, C., Wang, K., & van de Weijer, J. (2024) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme57554.2024.10687547 · Cited by 10
“To comprehensively evaluate the quality of inversion methods, we employ five metrics including MSE, LPIPS [37], SSIM [36], PSNR, and CLIP score [11] to verify the effectiveness of IterInv .”
MethodŠimka, M., Polák, L., Novotný, M., et al. (2024) · IEEE Access · doi:10.1109/access.2024.3478793 · Cited by 6
“The MS-SSIM [32] and VIFp [33] metrics were computed in MATLAB using an available script.”
MethodGeodesicPSIM: Predicting the Quality of Static Mesh With Texture Map via Geodesic Patch Similarity
Yang, Q., Jung, J., Xu, X., et al. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3501074 · Cited by 6
“The MSU Video Quality Measurement Tool [59] is used to compute the following IQA/VQA metrics: PSNR, SSIM [28], MSSIM [29], 3SSIM [60], VQM [61], and VMAF [30].”
InfluentialProcessing In-Memory PUF Watermark Embedding With Cellular Memristor Network
James, A. P., Reghuvaran, C., & Chua, L. O. (2025) · IEEE Transactions on Emerging Topics in Computing · doi:10.1109/tetc.2025.3528336 · Cited by 6
“Here,weuse C 1 = 0 and C 2 = 0 asin[37].”
Towards an Efficient Remote Sensing Image Compression Network with Visual State Space Model
Wang, Y., Liang, F., Wang, S., Chen, H., Cao, Q., Fu, H., & Chen, Z. (2025) · Remote Sensing · doi:10.3390/rs17030425 · Cited by 8
Bukka, V. V. R., Xu, M., Andrew, M., et al. (2025) · Methods in microscopy · doi:10.1515/mim-2024-0017 · Cited by 6
Exploring Invertible Encoding for Deep Video Compression
Guo, H., Kwong, S., & Zhou, M. (2025) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2025.3541869 · Cited by 6
“Moreover, for the MS-SSIM oriented model, we calculate the MS-SSIM [83] with luminance, contrast and structure comparison measures.”
GLNet: Global-Local Fusion Network for Strip Steel Surface Defects Detection
Wang, X., Bao, L., Zhou, X., Xia, L., & Xu, X. (2025) · IEEE Signal Processing Letters · doi:10.1109/lsp.2025.3546888 · Cited by 6
“[26], CEL [27], IOU [28] and SSIM [29].”
Pan, C., Jiang, Q., Zheng, H., Wang, P., Jin, X., Yao, S., & Zhou, W. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3548806 · Cited by 21
“…Metrics: The objective metrics chosen are entropy (EN) [51], standard deviation (SD) [52], spatial frequency (SF) [53], average gradient (AG) [54], sum of correlations of differences (SCDs) [55], gradient-based fusion performance (Qabf) [56], and multiscale structural similarity (MS-SSIM) [57].”
MB-RACS: Measurement-Bounds-Based Rate-Adaptive Image Compressed Sensing Network
Huang, Y., Chen, B., Li, N., An, B., Xia, S., & Wang, Y. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3549986 · Cited by 9
“Specifically, the peak signal-to-noise ratio (PSNR) provides a standard measure of reconstruction fidelity, while the multi-scale structural similarity (MS-SSIM) [43] offers insights into perceived image quality across different scales.”
Niu, X., Bai, B. B., Guo, N., Zhang, W., & Han, W. (2025) · Entropy · doi:10.3390/e27040373 · Cited by 9
“Distortion measures, such as mean squared error (MSE), structural similarity index (SSIM), its multiscale variant (MS-SSIM) [77], and peak signal-to-noise ratio (PSNR), require a ground-truth image for comparative analysis (see Figure 1c).”
Reconstruction and enhancement techniques for overcoming occlusion in facial recognition
Pleško, F., Goldmann, T., & Malinka, K. (2025) · EURASIP Journal on Image and Video Processing · doi:10.1186/s13640-025-00670-7 · Cited by 6
“…Index (SSIM) [9] metrics, Learned Perceptual Image Patch Similarity (LPIPS) [46], Feature Similarity Index (FSIM) [47], Multi-scale Structural Similarity (MS-SSIM) [48], Fréchet Inception Distance (FID) [49], which facilitate direct comparison between the proposed solution and existing methods.”
MicroSSIM: Improved Structural Similarity for Comparing Microscopy Data
Ashesh, A., Deschamps, J. R., & Jug, F. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-91721-9_16 · Cited by 14
HarmonyIQA: Pioneering Benchmark and Model for Image Harmonization Quality Assessment
Xu, Z., Duan, H., Ma, G., Liu, Y., Wang, J., Wu, Q., Min, X., Zhai, G., & Callet, P. L. (2025) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme59968.2025.11209987 · Cited by 17
Peculiarities of 3D compression of noisy multichannel images
Kozhemiakin, R. A., Абрамов, С. К., Лукин, В. В., Djurović, I., & Vozel, B. (2015) · Mediterranean Conference on Embedded Computing (MECO) · doi:10.1109/meco.2015.7181936 · Cited by 6
COHERENSi: A new full-reference IQA index using error spectrum chaos
Hegazy, T., & AlRegib, G. (2014) · IEEE Global Conference on Signal and Information Processing · doi:10.1109/globalsip.2014.7032264 · Cited by 5
“V.”
Dumić, E., Grgić, S., Jiménez, D., & Cruz, L. A. D. S. (2014) · International Symposium ELMAR (ELMAR) · doi:10.1109/elmar.2014.6923330 · Cited by 6
“So far several image and 2D video measures have been compared with the MOS (Mean Opinion Score) scores from the previously described 3D database: MS-SSIM (Multiscale Structural Similarity Index) [11], NQM (Noise Quality Measure) [12], PSNR (Peak Signal to Noise Ratio), SSIM (Structural Similarity Index) [13], VIF (Vis…”
MethodStereoscopic Image Quality Assessment Based on Cyclopean Image
Lu, K., & Zhu, W. (2016) · IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf on Cloud and Big Data Computing, Intl Conf on Cyber Science and Technology Congress (DASC/PiCom/CBDCom/CyberSciTech) · doi:10.1109/dasc-picom-datacom-cyberscitec.2016.169 · Cited by 6
“Such as Z. Wang.et proposed Structure Similarity (SSIM) [2] index and Mean Structural Similarity Index (MSSIM) [3] based on the idea of human visual system is sensitive with the structural information of image, A. Shnayderman.et proposed SVD-based Measure for local and global assessment (SVD) [4], and so on.”
BackgroundNaturalization of Screen Content Images for Enhanced Quality Evaluation
Guo, X., Huang, L., Gu, K., Li, L., Zhou, Z., & Tang, L. (2017) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2016edl8178 · Cited by 6
Mele, M. L., Millar, D., & Rijnders, C. E. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-58071-5_27 · Cited by 6
Improving point matching on multimodal images using distance and orientation automatic filtering
Anzid, H., Goïc, G. L., Bekkarri, A., et al. (2016) · doi:10.1109/aiccsa.2016.7945753 · Cited by 6
“The MSSIM measure is the mean of SSIM index [26] that”
Method
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.