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 1,201–1,300 of 6,047
Axial Super-Resolution Study for Optical Coherence Tomography Images Via Deep Learning
Yuan, Z., Yang, D., Pan, H., & Liang, Y. (2020) · IEEE Access · doi:10.1109/access.2020.3036837 · Cited by 29
“The MSSSIM [37] is calculated by combining the measurement at different scales.”
MethodImproving Image Quality of Single Plane Wave Ultrasound via Deep Learning Based Channel Compounding
Rothlübbers, S., Strohm, H., Eickel, K., et al. (2020) · IEEE International Ultrasonics Symposium · doi:10.1109/ius46767.2020.9251322 · Cited by 30
“The loss was computed as a linear combination of mean-squared error (MSE) and multiscale structural similarity (MS-SSIM) [9] loss on the log compressed, normalized final images as L = εLMSE + LMS−SSIM.”
MethodRemote sensing image super-resolution using cascade generative adversarial nets
Guo, D., Xia, Y. J., Xu, L., Li, W., & Luo, X. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.02.026 · Cited by 30
Emotional Semantics-Preserved and Feature-Aligned CycleGAN for Visual Emotion Adaptation
Zhao, S., Chen, X., Yue, X., Lin, C., Xu, P., Krishna, R., Yang, J., Ding, G., Sangiovanni-Vincentelli, A. L., & Keutzer, K. (2021) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2021.3062750 · Cited by 33
“According to [73], MS-SSIM is defined as”
MethodLearned Multi-Resolution Variable-Rate Image Compression With Octave-Based Residual Blocks
Akbari, M. A., Liang, J., Han, J., & Tu, C. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3068523 · Cited by 38
“It outperforms the BPG codec (RGB4:4:4) in both PSNR and MS-SSIM [15].”
MethodInfluentialHeterogeneous Face Recognition via Face Synthesis With Identity-Attribute Disentanglement
Yang, Z., Liang, J., Fu, C., Luo, M., & Zhang, X. (2022) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2022.3160595 · Cited by 33
“Specifically, MS-SSIM aims to measure the difference between structural information of images, which is related to the attributes of objects in the view and independent to extrinsic factors such as illumination, contrast, and noise [71].”
InfluentialEnd-to-End Optimized 360° Image Compression
Li, M., Li, J., Gu, S., et al. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3208429 · Cited by 30
“…traditional distortion metrics [17] includ-162 ing mean square error (MSE), peak signal-to-noise ratio (PSNR), structural similarity (SSIM) [18], and multi-scale structural similarity index (MS-SSIM) [19] are exploited to evaluate the difference between the source images and decoded images.”
MethodJoint Learning of Salient Object Detection, Depth Estimation and Contour Extraction
Zhao, X., Pang, Y., Zhang, L., & Lu, H. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3222641 · Cited by 43
“The total training loss can be written as: For the depth estimation, we follow the related works [42]– [44] to adopt the combination loss of the L1 and SSIM [45].”
MethodJoint Attention-Guided Feature Fusion Network for Saliency Detection of Surface Defects
Jiang, X., Yan, F., Lu, Y., Wang, K., Guo, S., Zhang, T., Pang, Y., Niu, J., & Xu, M. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3218547 · Cited by 48
“And ℓ iou and ℓ ssim denote BCE loss [30], IoU [31] loss and SSIM [32] loss, respectively.”
BackgroundQuality Evaluation of Arbitrary Style Transfer: Subjective Study and Objective Metric
Chen, H., Shao, F., Chai, X., Gu, Y., Jiang, Q., Meng, X., & Ho, Y. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3231041 · Cited by 41
“For the performance test on the CP, SR and OV quality evaluations, we compare the proposed SRQE with existing IQA metrics, including two categories: (1) Fourteen state-of-theart general-purpose FR-IQA metrics: SSIM [35], FSIM [26], MS-SSIM [57], IW-SSIM [25], Peak Signal-to-Noise Ratio (PSNR), MAD [58], VIF [24], VSI…”
BackgroundInfluentialYang, J., Wu, X., Bi, Z., & Geng, Z. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3264593 · Cited by 31
“Another loss function supervising the RGT estimation is based on RGT labels and we define it with the structural similarity (SSIM) metric [57].”
MethodNon-Semantics Suppressed Mask Learning for Unsupervised Video Semantic Compression
Tian, Y., Lu, G. S., Zhai, G., & Gao, Z. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01252 · Cited by 46
RawHDR: High Dynamic Range Image Reconstruction from a Single Raw Image
Zou, Y., Yan, C. C., & Fu, Y. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01133 · Cited by 59
“We can see that our method outperforms competing methods in PSNR, PSNR-µ , MS-SSIM, and presents a competitive performance in SSIM.”
MethodInfluentialGeneralizing Event-Based Motion Deblurring in Real-World Scenarios
Zhang, X., Yu, L., Yang, W. T., Liu, J., & Xia, G.-S. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00985 · Cited by 37
“Metrics PSNR and SSIM [28] are computed based on sequence restoration, i.e ., restoring 7 sharp images from one blurry input, for quantitative evaluation.”
MethodPredicting Chroma from Luma in AV1
Trudeau, L., Egge, N. E., & Barr, D. R. W. (2018) · DCC · doi:10.1109/dcc.2018.00046 · Cited by 30
“The BD-rate is measured using the following objective metrics: PSNR, PSNR-HVS [13], SSIM [14], CIEDE2000 [15] and MSSIM [16].”
MethodLearning to Distort Images Using Generative Adversarial Networks
Chen, L., Bampis, C. G., Li, Z., & Bovik, A. C. (2020) · IEEE Signal Processing Letters · doi:10.1109/lsp.2020.3040656 · Cited by 33
Self-citationSemantically Decomposing the Latent Spaces of Generative Adversarial NetworksPreprint
Donahue, C., Lipton, Z. C., Balsubramani, A., & McAuley, J. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.07904 · Cited by 120
“The multi-scale structural similarity (MS-SSIM) [39] metric reports the similarity of two images on a scale from 0 (no resemblance) to 1 (identical images).”
BackgroundTowards Perceptually Optimized End-to-end Adaptive Video StreamingPreprint
Bampis, C. G., Li, Z., Katsavounidis, I. X., Huang, T.-Y., Ekanadham, C., & Bovik, A. C. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.03898 · Cited by 106
“In our first experiment, we applied several well-known video quality and QoE metrics, including PSNR, PSNRhvs [63], SSIM [64], MS-SSIM [65], ST-RRED [66], VMAF [3] (version 0.6.1), SQI [10] and Video ATLAS [15].”
MethodInfluentialSelf-citationContextual Residual Aggregation for Ultra High-Resolution Image InpaintingPreprint
Yi, Z., Tang, Q., Azizi, S., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.09704 · Cited by 30
“Quantitative comparisons Table 2 reports our quantitative evaluation results in terms of mean L1 error, MSSSIM [27], Inception Score (IS) [22], and Frechet Inception Distance (FID) [14].”
Salehi, P., Chalechale, A., & Taghizadeh, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.13178 · Cited by 89
““Multi-Scale Structural Similarity for Image Quality” (MS-SSIM)[38] differs from the “Single-Scale Structural Similarity for Image Quality” (SS-SSIM)[39], which can be used to measure the similarity between two images.”
MethodImage Quality Assessment for Perceptual Image Restoration: A New Dataset, Benchmark and MetricPreprint
Gu, J., Cai, H., Chen, H., Ye, X., Ren, J. S. J., & Dong, C. (2020) · arXiv.org · doi:10.48550/arxiv.2011.15002 · Cited by 52
“, 2004), MS-SSIM (Wang et al., 2003), IFC (Sheikh et al.”
Data augmentation using CycleGAN-based methods for automatic bridge crack detection
Li, B., Guo, H., & Wang, Z. (2024) · Structures · doi:10.1016/j.istruc.2024.106321 · Cited by 39
Self-citationWu, J., Jing, R., Bai, Y., Tian, Z., Chen, W., Zhang, S., Yu, F. R., & Leung, V. C. M. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3386572 · Cited by 37
“2) Evaluation Metrics: We use the full-reference image quality assessment (IQA) metrics [46], PSNR, and structural similarity (SSIM) to quantitatively evaluate algorithms’ super-resolution reconstruction performance.”
MethodSIGFusion: Semantic Information-Guided Infrared and Visible Image Fusion
Lv, G., Sima, C., Gao, Y., Dong, A., Ma, G., & Cheng, J. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3457951 · Cited by 46
“…Evaluation Metrics: We select some objective evaluation metrics, including peak signal-to-noise ratio (PSNR) [53], correlation coefficient (CC) [54], Q AB / F [55], SSIM [51], multi-scale structural similarity index measure (MS-SSIM) [56], sum of correlation differences (SCD) [57], and FMI w [58].”
BackgroundMDVSC—Efficient Wireless Model Division Video Semantic Communication
Bao, Z., Liang, H., Dong, C., Li, C., Xu, X., & Zhang, P. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3464614 · Cited by 47
“The MDVSC is trained in terms of the mean square error (MSE) and tested under the PSNR, or MS-SSIM [28] for perceptual quality.”
Deep Learning-Based Semantic Communication System for Wireless Image Transmission
Hu, L.-L., Yu, L., & Qin, Z. (2025) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2025.3570499 · Cited by 29
“The performance of our proposed model is evaluated using peak signal-to-noise ratio (PSNR) and multi-scale structural similarity (MS-SSIM) metrics [27].”
USTC-TD: A Test Dataset and Benchmark for Image and Video Coding in 2020s
Li, Z., Liao, J., Tang, C., Zhang, H., Li, Y., Bian, Y., Sheng, X., Feng, X., Li, Y., Gao, C., et al. (2025) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2025.3608643 · Cited by 98
“7) Evaluative Metrics: For the objective quality evaluation, PSNR , MS-SSIM [39], and VMAF [40] are used to measure the quality of the coded frames in comparison to the original frames.”
Khan, M. I., Jeoti, V., & Khan, M. A. (2010) · International Conference on Intelligent and Advanced Systems · doi:10.1109/icias.2010.5716133 · Cited by 32
Structural similarity weighting for image quality assessment
Gu, K., Zhai, G., Yang, X., Zhang, W., & Liu, M. (2013) · IEEE International Conference on Multimedia and Expo workshops · doi:10.1109/icmew.2013.6618416 · Cited by 32
A dynamic system model of time-varying subjective quality of video streams over HTTP
Chen, C., Choi, L. K., Veciana, G. A. D., Caramanis, C., Heath, R. W., & Bovik, A. C. (2013) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2013.6638329 · Cited by 26
Self-citationBenchmarking of quality metrics on ultra-high definition video sequences
Hanhart, P., Korshunov, P., & Ebrahimi, T. (2013) · IEEE International Conference on Digital Signal Processing (DSP) · doi:10.1109/icdsp.2013.6622760 · Cited by 30
“In this study, the performance of the following objective metrics was assessed: 1) PSNR: Peak Signal-to-Noise Ratio, 2) VSNR: Visual Signal-to-Noise Ratio [10], 3) SSIM: Structural Similarity Index [11], 4) MS-SSIM: Multi-Scale Structural Similarity Index [12], 5) VIF: Visual Information Fidelity 1 [13], 6) VQM: Video…”
MethodPerceptual image quality assessment using block-based multi-metric fusion (BMMF)
Jin, L., Egiazarian, K., & Kuo, C. ‐. J. (2012) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2012.6288089 · Cited by 28
Blind high dynamic range image quality assessment using deep learning
Jia, S., Zhang, Y., Agrafiotis, D., & Bull, D. (2017) · International Conference on Information Photonics · doi:10.1109/icip.2017.8296384 · Cited by 28
“The FRIQA MSSSIM [14] and NR-IQA Marziliano [15] achieved better performance on the tone-mapped dataset.”
MethodHelmrich, C. R., Bosse, S., Siekmann, M., et al. (2019) · DCC · doi:10.1109/dcc.2019.00025 · Cited by 29
Adaptive Binarization of QR Code Images for Fast Automatic Sorting in Warehouse Systems
Chen, R., Yu, Y., Xu, X., et al. (2019) · Sensors · doi:10.3390/s19245466 · Cited by 29
“The definition is as follows [25]: l(x, y) = 2μ μ + C μ + μ + C , c(x,y) = 2σ σ + C σ + σ + C , s(x,y) = σ + C σ σ + C (9) The weights α, β and γ are greater than 0, which are parameters for adjusting the relative importance of the three comparison functions.”
Normalization of multicenter CT radiomics by a generative adversarial network method
Li, Y., Han, G., Wu, X., Li, Z. H., Zhao, K., Zhang, Z., Liu, Z., & Liang, C. (2020) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ab8319 · Cited by 31
“To overcome image deformation in unsupervised image translation, we combined the position-based selection strategy (Yang et al 2018) with multi-scale structure similarity loss (Wang et al 2003, Gokaslan et al 2018).”
Tao, L., Fisher, J., Anaya, E., Li, X., & Levin, C. S. (2020) · IEEE Transactions on Radiation and Plasma Medical Sciences · doi:10.1109/trpms.2020.2989073 · Cited by 31
Hao, Q., Zhou, K. Q., Yang, J., Hu, Y., Chai, Z., Ma, Y., Liu, G., Zhao, Y., Gao, S., & Liu, J. (2020) · Journal of Biomedical Optics · doi:10.1117/1.jbo.25.12.123702 · Cited by 30
“Compared with the single-scale structural similarity index, the MSSSIM supplies more flexibility in incorporating the variations of viewing conditions [23].”
MethodHe, Z., Jiang, G., Jiang, Z., et al. (2021) · Proceedings - International Conference on Image Processing · doi:10.1109/icip42928.2021.9506762 · Cited by 30
“[6] mapped colored PCs to 2D planes and used traditional image quality indicators (such as PSNR [7], SSIM [8], MSSSIM [9], and VIF [10]) to the resulting projection planes.”
MethodSignal Injection Attacks against CCD Image Sensors
Köhler, S., Baker, R., & Martinović, I. (2022) · ACM Asia Conference on Computer and Communications Security (ASIA CCS) · doi:10.1145/3488932.3497771 · Cited by 30
Self-Supervised Generative Style Transfer for One-Shot Medical Image Segmentation
Tomar, D., Bozorgtabar, B., Lortkipanidze, M., Rad, M. S., & Thiran, J. (2022) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv51458.2022.00180 · Cited by 34
“L cycle = LSSIM-L1(s,A(s̃, Estyle(s))) (3) where LSSIM-L1 computes multi-scale structural similarity index [46] and L1 distance between the two images as: LSSIM-L1(u, v) = ∥∥u− v∥∥ 1 + (1− SSIM(u, v)) (4)”
Rethinking Image Deblurring via CNN-Transformer Multiscale Hybrid Architecture
Zhao, Q., Yang, H., Zhou, D., & Cao, J. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3230482 · Cited by 42
“(VIF) [61], multiscale SSIM (MS-SSIM) [62], LPIPS [63], and DISTS [64].”
MethodOptimal Transport Guided Unsupervised Learning for Enhancing Low-Quality Retinal Images
Zhu, W., Qiu, P., Farazi, M., Nandakumar, K., Dumitrascu, O. M., & Wang, Y. (2023) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi53787.2023.10230719 · Cited by 31
“To preserve more local structures, particularly those that are pathologically meaningful, we choose Multi-Scale Structural Similarity Index Measure SSIMMS [11, 12] as our consistency loss, which is given by”
MethodJiang, G., Song, H., Yu, M., Song, Y., & Peng, Z. (2017) · IEEE Access · doi:10.1109/access.2017.2782320 · Cited by 31
“Yeganeh and Wang [5] therefore proposed a tone mapping quality index (TMQI) based on a combination of the modified multi-scale structural similarity (MS-SSIM) [6] and statistical naturalness.”
MethodQuality Assessment of Stereoscopic 360-degree Images from Multi-viewports
Xu, J., Luo, Z., Zhou, W., Zhang, W., & Chen, Z. (2019) · Picture Coding Symposium · doi:10.1109/pcs48520.2019.8954555 · Cited by 27
“For 2D metrics such as PSNR, SSIM [11], MS-SSIM [12], FSIM [27], S-PSNR [19], WS-PSNR [20], CPP-PSNR [21], the qualities of left and right view images are averaged to obtain the final perceptual score.”
Quality Prediction on Deep Generative Images
Ko, H., Lee, D. Y., Cho, S., & Bovik, A. C. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.2987180 · Cited by 31
“The thirteen existing models that were used for performance benchmarking are PSNR, SSIM [30], multi-scale SSIM index (MSSIM) [31], visual signal-to-noise ratio (VSNR) [58], visual information fidelity (VIF) [59], pixel-based VIF (VIFP), universal quality index (UQI) [60], information fidelity criterion (IFC) [61], noi…”
BackgroundMethodInfluentialSelf-citationCA-GAN: Weakly Supervised Color Aware GAN for Controllable Makeup Transfer
Kips, R., Gori, P., Perrot, M., & Bloch, I. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-67070-2_17 · Cited by 39
“The style transfer accuracy is then computed using the MSSIM similarity [35] as a measure of a perceptual distance.”
MethodNTIRE 2021 Challenge on Quality Enhancement of Compressed Video: Dataset and Study
Yang, R., & Timofte, R. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00076 · Cited by 33
“This may indicate that when optimizing the enhancement networks for PSNR, the MS-SSIM performance will be normally consistent with PSNR.”
MethodInfluentialA rasterized ray-tracer pipeline for real-time, multi-device sonar simulation
Cerqueira, R., Trocoli, T., Albiez, J., & Oliveira, L. (2020) · Graphical Models · doi:10.1016/j.gmod.2020.101086 · Cited by 32
“Four metrics were chosen to compute the similarity between the acoustic frames of the real and simulated images: Mean-squared error (MSE), peak signal-to-noise (PSNR), structural similarity index measure (SSIM) (Yang et al., 2008) and multi-scale structural similarity index measure (MS-SSIM) (Wang et al., 2003).”
BackgroundMethodInfluentialEnhanced Context Mining and Filtering for Learned Video Compression
Guo, H., Kwong, S., Ye, D., & Wang, S. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3316429 · Cited by 29
“Mean square error (MSE) or multi-scale structural similarity (MS-SSIM) [66] is applied to evaluate the distortion, respectively.”
MethodDenoising Diffusion Probabilistic Model for Face Sketch-to-Photo Synthesis
Que, Y., Li, X., Wan, W., Xia, X., & Liu, Z. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3409184 · Cited by 35
“To comprehensively assess the performance of the model in generating photo faces, we use the multiscale structural similarity metric (MSSIM) [49], Fréchet inception distance (FID) [50], spectral residual-based similarity (SR-SIM) [51], feature similarity (FSIM) [52], visual information fidelity…”
MethodInfluentialRobust Image Semantic Coding With Learnable CSI Fusion Masking Over MIMO Fading Channels
Xie, B., Wu, Y., Shi, Y., Zhang, W., Cui, S. R., & Debbah, M. A. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3409735 · Cited by 42
“Unless specifically marked, we choose multi-scale structural similarity (MS-SSIM) [45] for the basic reconstruction loss L 1 while setting MIMO antenna numbers as N T = 2 and N R = 2 .”
MethodLang, S.-J., Zhou, M.-L., Wei, X.-K., Yan, J.-L., Feng, Y., & Jia, W.-J. (2026) · IEEE transactions on broadcasting · doi:10.1109/tbc.2025.3609055 · Cited by 50
“…Indices: This work conducts a comparative analysis between hand-crafted feature-based methodologies, which include PSNR [16], SSIM [17], MS-SSIM [29], VIF [30], MAD [50], FSIM [56], VSI [31], GMSD [57], NLPD [19], and SCQI [20], and advanced deep learning-driven approaches, which include…”
A no-reference video quality assessment based on Laplacian pyramids
Zhu, K., Hirakawa, K., Asari, V. K., & Saupe, D. (2013) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2013.6738011 · Cited by 28
“The performance comparison in Table 3 shows our NR VQA is competitive with the “general-purpose” FR VQA algorithms (MS-SSIM [16], V-VIF [17], MOVIE [18]) listed in [14] for H.”
MethodInstantaneous Video Quality Assessment for lightweight devices
Liotta, A., Mocanu, D. C., Menkovski, V., Cagnetta, L., & Exarchakos, G. (2013) · International Conference on Advances in Mobile Computing and Multimedia (MoMM) · doi:10.1145/2536853.2536903 · Cited by 30
“The MS-SSIM index [20] is an extension over SSIM index.”
MethodZeng, K., Yeganeh, H., & Wang, Z. (2016) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2016.7532790 · Cited by 31
Self-citationUsman, M. A., Usman, M. R., & Shin, S. Y. (2017) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2017.10.007 · Cited by 30
Four-Dimensional Tomographic Reconstruction by Time Domain Decomposition
Nikitin, V., Carlsson, M., Андерссон, Ф., & Mokso, R. (2019) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2019.2898088 · Cited by 26
“Each SSIM index is computed with respect to formula (6) in [44] by extracting two-dimensional regions containing the particles, and comparing these regions to the ones from the ground truth model in Figure 1(a, b).”
MethodImproved hybrid layered image compression using deep learning and traditional codecs
Fu, H., Liang, F., Lei, B., Bian, N., Zhang, Q., Akbari, M. A., Liang, J., & Tu, C. (2020) · Signal Processing Image Communication · doi:10.1016/j.image.2019.115774 · Cited by 29
Image steganography based on LSB substitution and encryption method: adaptive LSB+3
Solak, S. B., & Altınışık, U. (2019) · Journal of Electronic Imaging · doi:10.1117/1.jei.28.4.043025 · Cited by 28
Photon-limited ptychography of 3D objects via Bayesian reconstruction
Nikitin, V., Aslan, S., Yao, Y., Biçer, T., Leyffer, S., Mokso, R., & Gürsoy, D. (2019) · OSA Continuum · doi:10.1364/osac.2.002948 · Cited by 28
Chatterjee, S., & Das, A. (2019) · Soft Computing · doi:10.1007/s00500-019-04635-7 · Cited by 30
“It is a metric used to determine the similarity between the two images (Wang et al. 2003).”
MethodAdaBoost neural network and cyclopean view for no-reference stereoscopic image quality assessment
Messai, O., Hachouf, F., & Seghir, Z. A. (2020) · Signal Processing Image Communication · doi:10.1016/j.image.2019.115772 · Cited by 34
E-DBPN: Enhanced Deep Back-Projection Networks for Remote Sensing Scene Image Superresolution
Yu, Y., Li, X., & Fu-xian, L. (2020) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2020.2966669 · Cited by 30
“Nevertheless, this loss function may generate perceptually unsatisfied images because it does not take the quality of the image into consideration [59], [60].”
BackgroundPSNR of Highest Distortion Region: An Effective Image Quality Assessment Method
Baig, M. A., Moinuddin, A. A., & Khan, E. (2019) · IEEE Uttar Pradesh Section International Conference on Electrical, Electronics and Computer Engineering (UPCON) · doi:10.1109/upcon47278.2019.8980171 · Cited by 30
“Mean-squared-error (MSE) [2], peak-signal-to-noise-ratio (PSNR) [1], [2], structural similarity index (SSIM) [3], multiscale SSIM (MS-SSIM) [4], feature similarity (FSIM) [5], etc.”
Composition-preserving deep approach to full-reference image quality assessment
Varga, D. (2020) · Signal, Image and Video Processing · doi:10.1007/s11760-020-01664-w · Cited by 23
“Wang et al. [37] proposed a multi-scale SSIM index, which is determined over multiple scales of the input image.”
BackgroundImproving the Performance of Image Fusion Based on Visual Saliency Weight Map Combined With CNN
Yan, L., Cao, J., Rizvi, S., et al. (2020) · IEEE Access · doi:10.1109/access.2020.2982712 · Cited by 28
“Therefore, to effectively quantify fusion strategies (of visual saliency, MSVD, and CNN) in our algorithm, we use three quality metrics, namely, Q 0 [47], multiscale structural similarity (MS_SSIM) [48], and sum of correlations of differences (SCD) [49].”
MethodA New Image Compression Algorithm Based on Non-Uniform Partition and U-System
Zhang, Y., Cai, Z., & Xiong, G. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.2992940 · Cited by 29
“In addition, by considering the significance of visual effects, we use Structural Similarity (SSIM) [32], a well-received metric for evaluating the quality of compressed images.”
MethodNonLocal Channel Attention for NonHomogeneous Image Dehazing
Metwaly, K., Li, X., Guo, T., & Monga, V. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00234 · Cited by 28
“The evaluation metrics used to quantify the performance are Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) [48].”
MethodDisentangled Representation Learning for Multiple Attributes Preserving Face Deidentification
Gong, M., Liu, J., Li, H., Xie, Y., & Tang, Z. (2020) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2020.3027617 · Cited by 31
Ketcha, M. D., Marrama, M., Souza, A., et al. (2021) · Journal of medical imaging · doi:10.1117/1.jmi.8.5.052103 · Cited by 28
Learning digital camera pipeline for extreme low-light imaging
Zamir, S. W., Arora, A., Khan, S., Khan, F. S., & Shao, L. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.04.076 · Cited by 41
“In this work, we utilize the multi-scale structural similarity measure (MS-SSIM) [42]: LMS-SSIM(ŷ,y) = 1− 1 N N∑ p=1 MS-SSIM(ŷp,yp).”
MethodInfluentialE2SRI: Learning to Super-Resolve Intensity Images From Events
Mostafavi, M., Nam, Y., Choi, J., & Yoon, K. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3096985 · Cited by 31
“We report the peak signal-to-noise ratio (PSNR) in decibels (log-arithmic scale), SSIM [38] as a fraction between zero (less similar) and one (fully similar), mean squared error (MSE), and the LPIPS as a metric to evaluate the similarity of the high-level features in two images (as the value…”
BackgroundDeLFIQE—A Low-Complexity Deep Learning-Based Light Field Image Quality Evaluator
Zhao, P., Chen, X., Chung, Y. Y., & Li, H. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3106113 · Cited by 29
Training a Task-Specific Image Reconstruction Loss
Mustafa, A., Mikhailiuk, A., Iliescu, D. A., Babbar, V., & Mantiuk, R. (2022) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv51458.2022.00010 · Cited by 45
“al [38] have studied visual quality of images produced by the image super-resolution, denoising and demosaicing algorithms using L2, L1, SSIM [33] and MS-SSIM [34] as loss functions.”
MethodInfluentialA Crypto-Steganography Approach for Hiding Ransomware within HEVC Streams in Android IoT Devices
Almomani, I., Alkhayer, A., & El‐Shafai, W. (2022) · Sensors · doi:10.3390/s22062281 · Cited by 37
“Multi-Scale SSIM index (MS-SSIM) [50] Evaluate the multi-scale structural similarity between the stego and cover frames.”
MethodNeural Distributed Image Compression with Cross-Attention Feature Alignment
Mital, N., Özyılkan, E., Garjani, A., & Gündüz, D. (2023) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv56688.2023.00253 · Cited by 31
“We evaluated the image quality performance of the models using multi-scale structural similarity index measure (MS-SSIM), which is widely reported to be a more realistic measure for human perception of image quality [35], in comparison with mean-squared error distortion.”
MethodJiang, B., Wang, J., Wu, Y., Wang, S., Zhang, J., Chen, X., Li, Y., Li, X., & Wang, L. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3261545 · Cited by 33
Su, H., Liu, Q., Yuan, H., et al. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3347638 · Cited by 28
“…cloud quality metrics SSIM p , MS - SSIM p , IW - SSIM p , V IFP p and PSNR P use the image quality assessment methods SSIM [28], [29], MS - SSIM [30], IW - SSIM [31], V IFP [32], and PSNR, respectively, to predict the quality of the point cloud from the average quality of six projection images.”
MethodNew image-quality measure based on wavelets
Dumić, E., Grgić, S., & Grgić, M. (2010) · Journal of Electronic Imaging · doi:10.1117/1.3293435 · Cited by 30
Shao, F., Zhang, Z., Jiang, Q., Lin, W., & Jiang, G. (2016) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2016.2628082 · Cited by 29
“These six IQMs include SSIM [3], MS-SSIM [4], FSIM [5], IW-SSIM [6], VIF [7], and GMSD [8].”
MethodRA Code: A Robust and Aesthetic Code for Resolution-Constrained Applications
Chen, C., Zhou, B., & Mow, W. H. (2017) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2017.2741472 · Cited by 28
“Denote the target perceptual quality as T which can be measured in terms of an objective image quality measure, such as MS-SSIM [29].”
BackgroundMethodFang, Y., Zeng, Y., Jiang, W., et al. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3053465 · Cited by 27
“(24), the global perceptual quality of the fused image can be obtained by an average operation Finally, the overall perceptual score is computed through the combination of scale-level quality scores using the multi-scale pooling method [52] where ω s is the weight assigned to the s - th scale.”
MethodDrees, L., Demie, D. T., Paul, M. R., Leonhardt, J., Seidel, S., Döring, T., & Roscher, R. (2023) · Plant Methods · doi:10.1186/s13007-024-01205-3 · Cited by 27
“For the direct comparison between generated and reference images of the same time point, we use the Multi-scale Structural Similarity Index Measure (MS-SSIM [35], optimal: 1) and the Learned Perceptual Image Patch Similarity (LPIPS [36], optimal: 0).”
MethodGenerative Latent Coding for Ultra-Low Bitrate Image Compression
Jia, Z., Li, J., Li, B., Li, H., & Lu, Y. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.02465 · Cited by 95
“Additionally, we include PSNR and MS-SSIM [57] for completeness.”
Deep Learning to Simulate Contrast-Enhanced MRI for Evaluating Suspected Prostate Cancer
Huang, H., Mo, J., Ding, Z., Peng, X., Liu, R., Zhuang, D., Zhang, Y., Hu, G., Huang, B., & Qiu, Y. (2025) · Radiology · doi:10.1148/radiol.240238 · Cited by 30
Shao, F., Tian, W., Lin, W., Jiang, G., & Dai, Q. (2017) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2017.2685240 · Cited by 28
“Too little or too much IQMs are not efficient to represent the quality space, as a compromise, we select six IQMs, including PSNR, SSIM [1], MS-SSIM [2], gradient magnitude (GM) from FSIM [4], GMSD [5] and VSI [8] (combination of the six IQMs will achieve comparatively better performance for the MDSIs, as compared with…”
BackgroundMethodInfluentialSubjective and objective quality assessment of omnidirectional video
Queluz, M. P., Lopes, F., Ascenso, J., & Rodrigues, A. (2018) · SPIE Applications of Digital Image Processing · doi:10.1117/12.2321679 · Cited by 26
Image Quality Assessment: A Review to Full Reference Indexes
Khosravy, M., Patel, N. V., Gupta, N., & Sethi, I. K. (2018) · Lecture notes in electrical engineering · doi:10.1007/978-981-13-2685-1_27 · Cited by 28
Wang, Y., Li, Q., & Chen, B. (2019) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2019.102647 · Cited by 27
Multi-Type Self-Attention Guided Degraded Saliency Detection
Zhou, Z., Wang, Z., Lu, H., Wang, S., & Sun, M. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i07.7010 · Cited by 32
“L total is expressed as: where and SSIM () (Wang, Simoncelli, and Bovik 2003) is the standard structural similarity function. λ 1 , λ 2 and λ 3 is set to 5, 10 and 2 respectively.”
MethodSelf-citationDeng, M., Goy, A., Li, S., Arthur, K., & Barbastathis, G. (2019) · Optics Express · doi:10.1364/oe.381301 · Cited by 28
A convolutional neural network‐based model observer for breast CT images
Kim, G., Han, M., Shim, H., et al. (2020) · Medical Physics · doi:10.1002/mp.14072 · Cited by 27
Learned Variable-Rate Image Compression With Residual Divisive Normalization
Akbari, M. A., Liang, J., Han, J., & Tu, C. (2020) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme46284.2020.9102877 · Cited by 34
“We use both PSNR and MS-SSIM [9] as the evaluation metrics.”
Visual Quality Assessment for Perceptually Encrypted Light Field Images
Wen, W., Wei, K., Fang, Y., & Zhang, Y. (2020) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3026817 · Cited by 28
Katzmann, A., Taubmann, O., Ahmad, S., Mühlberg, A., Sühling, M., & Groß, H. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.05.081 · Cited by 34
“Structural dissimilarity considers the image context, luminance, contrast and structure, and —from a perceptual point of view—is thus preferable to other loss functions such as L1 or binary cross entropy [49].”
BackgroundTowards Viewport-dependent 6DoF 360 Video Tiled Streaming for Virtual Reality Systems
Jeong, J.-B., Lee, S., Ryu, I.-W., et al. (2020) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3394171.3413712 · Cited by 27
Color Image Restoration Exploiting Inter-Channel Correlation With a 3-Stage CNN
Cui, K., Boev, A., Alshina, E. A., & Steinbach, E. G. (2020) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2020.3043148 · Cited by 32
“In order to ensure that this is a general characteristic for noisy images, we evaluate the average PSNR, MS-SSIM (Multi-scale Structural Similarity Index) [54], and noise variance for the red, green, and blue channels of the whole dataset.”
MethodAn unsupervised reconstruction method for low-dose CT using deep generative regularization prior
Unal, M. O., Ertaş, M., & Yıldırım, İ. (2022) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2022.103598 · Cited by 30
“To compare different options, the channel depths of three different network architectures are defined as follows: • v1 : 5 scales, [16 , 32 , 64 , 128 , 256] .”
BackgroundMethodInfluentialPIQI: perceptual image quality index based on ensemble of Gaussian process regression
Ahmed, N., Asif, H. M. S., & Khalid, H. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-020-10286-w · Cited by 30
“We have explored the role of multi-scale [41] and multi-color space [40, 42, 50] in the extraction 15678”
BackgroundMethodInfluential
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.