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 5,501–5,600 of 6,047
Occlusion Handling Algorithm Based on Contour Detection
Dai, Z., Hu, X., Chen, C., & Yu, H. (2024) · Journal of Advanced Computational Intelligence and Intelligent Informatics · doi:10.20965/jaciii.2024.p0893
Ramsook, D., Vibhoothi, V., Kokaram, A. C., et al. (2024) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex61742.2024.10598241
“Recent research in measurement of the subjective quality of neural based approaches trained on different optimization criteria [18], demonstrated that participants preferred content that was restored using either DISTS [19] or MS-SSIM [20] as the optimization criteria.”
Enhanced Unsupervised Ultrasound Video Object Segmentation through Multi-scale Feature Fusion
Tawhidi, J., Wang, M., & Wen, Y. (2024) · International Conference on Computing, Networks and Internet of Things (CNIOT) · doi:10.1145/3670105.3670157
51‐2: Invited Paper: Foveated Image Transmission with Anti‐Aliasing Image Reconstruction
Tzung-Yuan, L., Yu-Hsing, C., Yao-Ming, L., et al. (2024) · SID Symposium Digest of Technical Papers · doi:10.1002/sdtp.17620
47‐2: Invited Paper: Modulated Analog Driving and Evaluation of Image Quality
Henzen, A., Rockoff, T. E., & Caulkins, W. (2024) · SID Symposium Digest of Technical Papers · doi:10.1002/sdtp.17602
Learned Image Compression with High-Frequency Key Information Extraction and Enhancement
Zhao, Z., & Zhao, Y. (2024) · International Conference on Computer and Communication Systems (ICCCS) · doi:10.1109/icccs61882.2024.10603055
mohammadi, M. M., Raeisi, Z., & Lashaki, R. A. (2024) · Research Square · doi:10.21203/rs.3.rs-4716351/v1
A Quality Metric for Semantically Transmitted Images in Machine-to-Machine CommunicationsPreprint
Gowrisetty, V. S. S., Lokumarambage, M., Samarathunga, P., et al. (2024) · Research Square · doi:10.21203/rs.3.rs-4616580/v1
Blood Harmonisation of Endoscopic Transsphenoidal Surgical Video Frames on Phantom Models
Saeed, M., Quarez, J., Irzan, H., et al. (2024) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi56570.2024.10635809
Voggu, A. R., & Rao, M. (2024) · International Conference on Signal Processing and Communications (SPCOM) · doi:10.1109/spcom60851.2024.10631588
“DSSIM is a Multi-Scale variant [28] of SSIM [29] which models Human Vision to provide image similarity measurements with a better accuracy compared to SSIM.”
MethodLow Bit Rate Video Coding using VVC and DCVC-DC for River Surveillance
Morimoto, Y., Katayama, T., Song, T., & Shimamoto, T. (2024) · International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC) · doi:10.1109/itc-cscc62988.2024.10628310
Liu, J., Yan, J., Han, J., Chen, C., Wang, X., & Miao, C. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3449279
“(SSIM) to evaluate generators [51], [52], [53], [54].”
GenPower-NeRF: A Neural Radiance Field Method with Powerful Generalization
Ding, Z., Zhang, Z., & Liang, Y. (2024) · doi:10.1109/ijcnn60899.2024.10651402
NeRF-SR++: Towards Higher Quality Supersampled Neural Radiation Fields
Xiang, Q., Xiao, J., Yu, W., et al. (2024) · doi:10.1109/ijcnn60899.2024.10650329
“Restrictions apply. metrics: Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) [29] and LPIPS [30].”
BackgroundEfficiency Across Datasets: Insights into Video Compression Techniques
Sreeram, A., Chandra, B., Swetha, G., & Bagade, S. (2024) · International Conference on Image Processing and Capsule Networks (ICIPCN) · doi:10.1109/icipcn63822.2024.00011
“Evaluation: Bitrates were assessed, and a distortion metre for RGB color [13] space was employed.”
MethodA Shallow and Deep Feature Fusion Model for Image Quality Assessment
Liang, L., & Xue, C. (2024) · Chinese Control Conference (CCC) · doi:10.23919/ccc63176.2024.10661958
Generalized dithering using the lattice Boltzmann method
Noël, R., Renier-Robin, A., & Navarro, L. (2024) · Signal Image and Video Processing · doi:10.1007/s11760-024-03465-x
“In its generic formulation, the SSIM can include these properties measurement with different weights [21].”
Artificial Intelligence Model for 3D Seismic Fault Analysis
Liao, Z. (2024) · Springer eBooks · doi:10.1007/978-981-97-7079-3_11
3DST: A Robust End-to-End 3D Scene Transmission Framework Using Neural Radiance Fields
Zhang, Y., Yue, W., Niu, K., et al. (2024) · IEEE/CIC International Conference on Communications in China (ICCC) · doi:10.1109/iccc62479.2024.10681998
“Evaluation metrics: In order to evaluate rendering quality, we qualify the performance of the proposed method and image transmission scheme using both the widely used pixel-wise metric (PSNR [24]) and the perceptual metric (MS-SSIM [17]).”
MethodQuality of Experience of Viewport Adaptive Omnidirectional Video Streaming
Liu, X., Zhang, H., Yan, J., Zhang, H., Fang, Y., & Wang, S. (2024) · Proceedings - International Conference on Image Processing · doi:10.1109/icip51287.2024.10647716
Lightweight Foreground Segmentation with the Visual Mamba Model
Qi, X., & Jiang, Y. (2024) · doi:10.1145/3691016.3691045
“For the foreground mask supervision, we use a combination of BCE loss [ 3 ] , SSIM loss [ 4 ] , and IoU loss [ 5 ] to ensure our model focuses more on the pixels in the foreground areas and maintains visual integrity.”
Spatiotemporal satellite image fusion using nanosatellite data
Kim, Y., & Kim, H. O. (2024) · doi:10.1117/12.3029007
“Loss function was composed of mean squared error and multi-scale structural similarity index measure (Wang et al, 2003).”
Choudhury, P., & Boruah, B. R. (2024) · Journal of Microscopy · doi:10.1111/jmi.13362
Accelerating multi-coil MR image reconstruction using weak supervision
Atalık, A., Chopra, S., & Sodickson, D. K. (2024) · Magnetic Resonance Materials in Physics Biology and Medicine · doi:10.1007/s10334-024-01206-2
“Fully supervised fine-tuning and baselines For supervised fine-tuning and supervised baselines, we utilized the MS-SSIM loss function [29] with its default parameters, which employs a uniform kernel of 7 × 7 size and k values of 0.01 and 0.03, across 5 scales with the respective weights of 0.0448,…”
InfluentialYan, Z., Liu, P., Wang, X., et al. (2024) · Symmetry · doi:10.3390/sym16101342
Zhang, H., Jin, W., & Shi, Y. (2024) · doi:10.1117/12.3038602
Zhang, Y., Du, Y., & Wang, Y. (2024) · IEEE International Conference on Software Quality, Reliability and Security Companion (QRS-C) · doi:10.1109/qrs-c63300.2024.00084
Wen, W., Huang, M., Dong, L., et al. (2024) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2024.3495515
Li, Y., Li, X., Luo, J. S., et al. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.5007054
Neural light field reconstruction from hybrid cameras
Yu, K., Shi, L., Liu, C., et al. (2024) · doi:10.1117/12.3036229
Integrating Optical Flow into Deep Learning based Distortion Correction
researcher, I., Molnár, S.-B. L., & Pável, S. (2024) · Acta Universitatis Sapientiae Informatica · doi:10.47745/ausi-2024-0005 · Cited by 1
Parallax Compensation-Driven Feature Transform Network for Super-Resolution Image Quality Assessment
Yang, Y., Wei, X., Zhang, T., Shang, Z., Pu, H., Luo, J., Song, S., Ouyang, L., Mao, Q., & Zhou, M. (2024) · Journal of Circuits Systems and Computers · doi:10.1142/s0218126625501610
Similar Image Avoidance Policy Based on Image Similarity Measurement for Next Best View
Wang, J., Lee, W., Winata, I. M. P. A., et al. (2024) · International Conference on Control, Automation and Systems (ICCAS) · doi:10.23919/iccas63016.2024.10773048
Direct Diffusion Bridge for High MB-count Image Generation in Ultrasound Localization Microscopy
Zhang, S., Lan, H., Wang, R., Wei, X., Chen, Z., Liu, Y., He, Z., & Luo, J. (2024) · IEEE Ultrasonics, Ferroelectrics, and Frequency Control Joint Symposium (UFFC-JS) · doi:10.1109/uffc-js60046.2024.10793898
“In this work, normalized cross correlation (NCC), vessel area density (VAD), MS-SSIM [14] and Dice similarity coefficient (DSC) were used to evaluate the performance of the proposed method. denote the standard deviations, is the covariance of the two images.”
Face parsing and blending-based analysis model for makeup style transferPreprint
Wang, X., Zhang, Y., Han, T., et al. (2024) · Research Square · doi:10.21203/rs.3.rs-5642552/v1
Towards Neural Codec-Empowered 360$^\circ$ Video Streaming: A Saliency-Aided Synergistic Approach
Shi, J., Zhang, M., Shen, L., et al. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3521770
“Similarresultscanalsobeobservedintermsofother metrics [21], [22], [23], such as perceptually based SSIM [43] and LPIPS [44].”
Subset-Selection Weight Post-Training Quantization Method for Learned Image Compression Task
Yang, J., Wang, X., Li, Q., Qiao, S., & Zhou, Y. (2025) · IEEE Access · doi:10.1109/access.2024.3524553
“Their pre-trained models were trained using MSE loss and MS-SSIM [30] loss, respectively.”
Paskaš, M. P. (2024) · Telecommunications Forum (TELFOR) · doi:10.1109/telfor63250.2024.10819050
Tian, Y., Liu, M., & Su, D. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5081491
Fang, Z., Zhong, Q., Chen, H., et al. (2025) · doi:10.5194/egusphere-2024-2888-rc1
Lightweight Multi-Scale and Multi-Level Feature Fusion Network For Rgb-D-T Salient Object DetectionPreprint
Zhou, X., Cui, F., Bai, Z., Liu, D., Zhang, G., & Zhang, J. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5092774
Fang, Z. (2025) · doi:10.5194/egusphere-2024-2888-ac1
Selecting projection views based on error equidistribution for computed tomography
Zhang, Y., Zhao, X., Chen, K., et al. (2025) · Journal of X-Ray Science and Technology · doi:10.1177/08953996241289267
“35 When needed, Poisson noise speci fi ed by the incident intensity, denoted as I 0 , will be introduced to the projection data p , i.e. where p and p noisy denote noise-free and noisy projection data, respectively.”
Fang, Z. (2025) · doi:10.5194/egusphere-2024-2888-ac2
Mosaic Reduction Model Using Local Thermal Diffusion Equation
Lv, H. (2025) · IEEE Access · doi:10.1109/access.2025.3531005
Image Quality Assessments by Leveraging Diverse Visual Tasks
Lee, J., Park, D.-W., & Chun, S. Y. (2025) · IEEE Access · doi:10.1109/access.2025.3531500 · Cited by 1
FSDN: Image frequency and semantic decomposition network for image dehazing
Tong, Z., Liu, M., & Jin, X. (2024) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip63160.2024.10849875
Tests of Solar X-Ray Image Reconstruction: A New Index for Assessing Image Quality
Li, Z., Yu, W., Su, Y., et al. (2025) · Research in Astronomy and Astrophysics · doi:10.1088/1674-4527/adb13a
Perceptual visual security index: Analyzing image content leakage for vision language models
Hu, L., Xiang, T., Guo, S., Li, X., & Yang, Y. (2025) · Journal of Information Security and Applications · doi:10.1016/j.jisa.2025.103988 · Cited by 1
Distortion Correction for Wide-Angle Videos of Power Inspection Robots
Ren, Z., & Su, Y. (2024) · doi:10.1109/epee63731.2024.10875060
Asymmetric Multi-Layer Compression: Decoupling Inter-Layer Coding Dependencies With a Learned Model
Cancellier, L. H., Bräscher, A. B., Seidel, I., et al. (2025) · IEEE Access · doi:10.1109/access.2025.3542979
“…[27]; the PSNR-HVS-M [32], which is still based on PSNR but takes the Human Visual System (HVS) characteristics into consideration (frequency and contrast); we use the MS-SSIM metric [33], which also tries to improve the correlation with the HVS by considering the image structural information.”
A Data Enhancement and Fault Diagnosis Method for Planetary Gearboxes With Extremely Small Samples
Huang, T., Peng, C., Sheng, Y., et al. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3541226
“The SSIM index [21], on the other hand, quantifies the similarity between two images in terms of brightness, contrast, and structure, with values ranging from [ − 1, 1].”
Expert System to the Image Quality Assessment in Diffraction Problem with Linearly Polarized Beam
Khoroshun, A., Popiołek-Masajada, A., Riazantsev, A., et al. (2025) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-3-031-81557-7_15
ConvCycleGAN - An Unsupervised Generative Adversarial Network for Chinese Pastoral Painting Style
Dong, Z., He, N., Li, W., et al. (2024) · International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) · doi:10.1109/cisp-bmei64163.2024.10906297
“This section introduces the adversarial loss, cycle-consistency loss, identity loss, and MS-SSIM [19] loss used in both models.”
Kaderuppan, S. S., Sharma, A., Saifuddin, M. R., et al. (2024) · IEEE Region 10 Conference (TENCON) · doi:10.1109/tencon61640.2024.10902742
“…[e.g. PSNR [9] & image MSE (IMMSE) / image signature based mean square error (ISMSE) [10])] and the similarity between 2 images (SSIM [11], MS-SSIM [12] & FSIM C [13]) – 2 aspects which are particularly essential when evaluating the performance of a DCNN-generated image in the context of image…”
Local Adaptive Blur Attention-based Neural Network for Image Deblurring
Lee, M., Lee, H.-S., & Lee, H. (2025) · The Journal of Korean Institute of Information Technology · doi:10.14801/jkiit.2025.23.2.65
Post-Filtering of Noisy Images Compressed by HEIF
Kryvenko, S., Rebrov, V., Лукин, В. В., Golovko, V. A., Sachenko, A., Шелестов, А. Ю., & Vozel, B. (2025) · Applied Sciences · doi:10.3390/app15062939 · Cited by 3
Novel Model of Medical CT Image Segmentation Based on GANs With Residual Neural Networks
Bouden, A., Blaiech, A. G., Abdallah, A. B., et al. (2025) · International Journal of Imaging Systems and Technology · doi:10.1002/ima.70049
ThicknessVAE: Learning a Lateral Prior for Clothed Human Body Reconstruction
Wu, X., Fan, Z., He, H., et al. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10889190
A Low-Complexity Perceptual Video Quality Metric for SVT-AV1
Sivashanmugam, S., Decker, A. D., & Cock, J. D. (2025) · Mile-High Video Conference (MHV) · doi:10.1145/3715675.3715850
Huang, H., Balaji, S., Aslan, B., Wen, Y., Selim, M. H., Thomas, A. J., Filippidis, A., Spincemaille, P., Wang, Y. D., & Soman, S. (2025) · International Journal of Imaging Systems and Technology · doi:10.1002/ima.70070 · Cited by 1
“…(DICE) [28], mean normalized mutual information (NMI) [29], mean structural similarity (SSIM) [30], mean multi-scale structural similarity (MS-SSIM) [31], mean peak signal- to-noise ratio (PSNR) [32], mean information theoretic-based statistic similarity measure (ISSM) [33] (see Table 1).”
Pirate: No Compromise Low-Bandwidth VR Streaming for Edge Devices
Zhang, Y., Kang, Y., Ying, Z., et al. (2025) · ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) · doi:10.1145/3676641.3716268
“To assess image quality numerically, we have adopted two commonly used image quality metrics: PSNR (Peak Signal-to-Noise Ratio) and MS-SSIM (Multi-Scale Structural Similarity Index) [42].”
Perceptual Quality Driven Point Cloud Compression for 6DoF 3D Point Cloud Streaming
Chujo, Y., Tagashira, Y., Harada, Y., et al. (2024) · International Symposium on Multimedia · doi:10.1109/ism63611.2024.00034
“In this case, we use the MS-SSIM metric [12].”
Validation of Map Production through Deep Learning-based Satellite Image Super-Resolution
Park, J., Shin, S.-H., Kwak, S., et al. (2025) · Journal of the Korea Institute of Military Science and Technology · doi:10.9766/kimst.2025.28.2.155
“Wang [7] 은 [8] 이 제안한 방법인 L1 Loss 와 MS-SSIM [9] 을 각각 0.78, 0.22 가중치를 곱하여 사용하였다 .”
InfluentialEfficient lunar lander detection using a lightweight saliency map for small planetary rovers
Sakamoto, K., Minegishi, R., Maeda, T., et al. (2025) · ROBOMECH Journal · doi:10.1186/s40648-025-00300-x
“Based on the [34], the SSIM is calculated by: where µ x and µ y are the mean values of x and y , respectively, σ x and σ y are the standard deviations of x and y , respectively, σ xy is the covariance of x and y , and c 1 and c 2 are constant value to stabilize the division, respectively.”
Verdicchio, M., Grassi, F., Esposito, G., Cavaliere, C., Alfano, V., Cerrone, M., Pappatà, S., Salvatore, E., Cecca, A. D., Salvatore, M., et al. (2025) · Scientific Reports · doi:10.1038/s41598-025-87224-z
“For this study, the following measures have been evaluated: The Structural Similarity (SSIM) Index 32 , is a perceptual metric that assesses image quality degradation resulting from various processing techniques or data transmission losses.”
Implicitly Learned Neural Phase Functions for Point Spread Function Engineering
Valouev, A. (2024) · doi:10.1109/icvisp64524.2024.10959478
“To evaluate our model's performance, we used the Multiscale Structural Similarity (MSSIM) [12] and peak signal-to-noise ratio (PSNR) as metrics.”
Multi-Scale Receptive Field Architecture for Consumer-Grade Smartphone Image InpaintingPreprint
Wang, M., Chen, S., Lin, Z., et al. (2025) · doi:10.36227/techrxiv.174494580.00753592/v1
Guo, X., Ye, C., Zhang, M., Hao, X., Yang, Y., Yu, Y., Ma, T., & Han, Y. (2025) · International Journal of Imaging Systems and Technology · doi:10.1002/ima.70074
“In this study, the quality of synthesized T1 images is assessed using three metrics: mean absolute error (MAE), peak signal- to-noise ratio (PSNR), and multi-scale structural similarity (MS-SSIM) (Wang et al. [38]).”
One-click image reconstruction in single-molecule localization microscopy via deep learningPreprint
Saguy, A., Xiao, D., Narayanasamy, K. K., Nakatani, Y., Gustavsson, A., Heilemann, M., & Shechtman, Y. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.04.13.648574
End-to-End Learning Framework Incorporating Image Reconstruction and Recognition Models
Shibata, R., & Yamauchi, Y. (2025) · IEEE Access · doi:10.1109/access.2025.3563476
“Image reconstruction models designed for human vision, which use SSIM [11], MS-SSIM [12], and LPIPS [13] as loss functions and optimized them to achieve advanced image compression and restoration.”
Zhang, W., Wang, X., Lin, H., Xing, W., Zhao, L., & Yang, F. (2025) · IEEE Conference on Telecommunications, Optics and Computer Science (TOCS) · doi:10.1117/12.3067557
“They are multi-scale structural similarity (SSIM)[37], Inception Score (IS)[38], and Fréchet Inception Distance (FID)[39].”
Seeing Eye to AI? Applying Deep-Feature-Based Similarity Metrics to Information Visualization
Long, S., Chatzimparmpas, A., Alexander, E., Kay, M., & Hullman, J. R. (2025) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713955 · Cited by 3
“In other words, 𝑤 ,𝑙 ∈ L constitute a “perceptual calibration” of a few parameters in an existing feature space [87].”
InfluentialA Unique Approach for Extraction of Images with 4 Amplification Factors Using SR -GAN
Chaudhary, H., Kaur, S., Sharma, P., et al. (2024) · doi:10.1109/icpids65698.2024.00056
Implementation of neural network based 2D seismic images super resolution approach
Noskov, O., & Myrontsov, M. (2025) · Environmental Safety and Natural Resources · doi:10.32347/2411-4049.2025.1.139-145
Kong, F., Shi, Z., Ge, T., et al. (2025) · Journal of Instrumentation · doi:10.1088/1748-0221/20/05/p05004
“This hybrid loss function primarily incorporates the Mean Absolute Error (MAE) and Multi-scale Structural Similarity Index (MS-SSIM) metrics [35].”
Structurally Consistent MRI Colorization using Cross-Modal Fusion Learning
Mathur, M., Chaudhary, A., Gupta, S. K., & Sharma, O. (2025) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi60581.2025.10981112
An ‘Unsupervised’ 3D Image Quality Assessment using Spectral Decompositions of Scene Components
Kajrolkar, S., Madana, V., & Appina, B. (2025) · National Conference on Communications · doi:10.1109/ncc63735.2025.10983225
“SSIM [2], MS-SSIM [3] and VIF [4] are 2D FR IQA models, NIQE [20], PIQE [26] and ILNIQE [27] are unsupervised 2D NR IQA models.”
Bounded Editing: Multi-Object Image Manipulation with Region-Specific Control
Kang, M. G., Kim, K., & Choi, Y. S. (2025) · ACM Symposium on Applied Computing · doi:10.1145/3672608.3707793
Atjo: Adaptive Three-Dimensional Joint Optimization for Remote Sensing Video Super-ResolutionPreprint
Bu, L., Qin, T., Zhang, Z., et al. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5264262
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0010
Learning to solve inverse problems end-to-end
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0006 · Cited by 1
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0013
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0005
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0001
Deep Learning for Computational Imaging
Heckel, R. (2025) · doi:10.1093/oso/9780198947172.001.0001 · Cited by 4
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.002.0006
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0007
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.005.0001
Solving inverse problems with optimization tasks
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0002
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.002.0003
Datasets and evaluation of image reconstruction methods
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0014
Coordinate-based multi-layer perceptrons
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0012
Signal reconstruction with un-trained neural networks
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0011
Advanced reconstruction problems
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0015
Sparse modelling and sparse signal reconstruction
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0004
Signal reconstruction by imposing generative priors
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0009
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0008 · Cited by 1
Heckel, R. (2025) · Oxford University Press eBooks · doi:10.1093/oso/9780198947172.003.0003
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.