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 601–700 of 6,047
CLIPascene: Scene Sketching with Different Types and Levels of Abstraction
Vinker, Y., Alaluf, Y., Cohen‐Or, D., & Shamir, A. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00383 · Cited by 127
“Geometry preservation of our generated sketches by fidelity level, across different fidelity levels using the MSSSIM [45] metric.”
MethodGraph-Represented Distribution Similarity Index for Full-Reference Image Quality Assessment
Shen, W., Zhou, M.-L., Luo, J., Li, Z.-G., & Kwong, S. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3390565 · Cited by 66
“The multiscale SSIM (MS-SSIM) [8] incorporates more flexibility into the variations exhibited by observation conditions to calibrate the parameters that are relatively important at different scales.”
BackgroundMethodInfluentialEnhancing the Image Quality via Transferred Deep Residual Learning of Coarse PET Sinograms
Hong, X., Zan, Y., Weng, F., Tao, W., Peng, Q., & Huang, Q. (2018) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2018.2830381 · Cited by 67
“index (MS-SSIM) [41], and peak signal-to-noise ratio (PSNR).”
BackgroundDeep Objective Quality Assessment Driven Single Image Super-Resolution
Yan, B., Bare, B., Ma, C., Li, K., & Tan, W. (2019) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2019.2914883 · Cited by 64
“many other FR-IQA methods [43]–[45] have been developed based on the ideas of SSIM.”
MethodDeep Stereoscopic Image Super-Resolution via Interaction Module
Lei, J., Zhang, Z., Fan, X., Yang, B., Li, X., Chen, Y., & Huang, Q. (2020) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3037068 · Cited by 65
ImUnity: A generalizable VAE-GAN solution for multicenter MR image harmonization
Cackowski, S., Barbier, E., Dojat, M., & Christen, T. (2023) · Medical Image Analysis · doi:10.1016/j.media.2023.102799 · Cited by 86
Hyperspectral Data Compression Using Fully Convolutional Autoencoder
Grassa, R. L., Re, C., Cremonese, G., & Gallo, I. (2022) · Remote Sensing · doi:10.3390/rs14102472 · Cited by 78
Lou, A., Guan, S., & Loew, M. H. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.106579 · Cited by 95
“There are many popular methods to compare two images: • Binary to binary: Jaccard Similarity (JS) [37] • Gray to gray: Mean Absolute Error (MAE) [38]; Tanimoto Similarity [39] (extended Jaccard Similarity); Structural Similarity (SSIM) [40].”
MethodZhou, Q., He, S., Liu, H., Chen, T., & Chen, J. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3218587 · Cited by 88
“Chen et. al . extended this idea by combining multi-scale structural similarity index MS-SSIM [35] and mean absolute error to accurately identify structural and non-structural multi-scale anomalies.”
MethodA three-dimensional virtual mouse generates synthetic training data for behavioral analysis
Bolaños, L. A., Xiao, D., Ford, N. L., LeDue, J. M., Gupta, P. K., Doebeli, C., Hu, H., Rhodin, H., & Murphy, T. H. (2021) · Nature Methods · doi:10.1038/s41592-021-01103-9 · Cited by 75
“We observed that the SSIM scores for the paired (same mouse) synthetic-real images were higher than for both the unpaired mouse sets and the aggregate model (mean = 0.543).”
MethodInfluentialDeep data analyzing algorithm based on scale space theory
Zhu, Y., & Gao, K. (2016) · Cluster Computing · doi:10.1007/s10586-016-0677-3 · Cited by 64
“Based on the knowledge of scale transformation of the definition and study field scale conversion classification [5], from knowledge of themultiscale conversion direction angle, multiscale data analysis algorithm scale on Pushover analysis algorithm and scale pushdown analysis algorithm: (1) Scale up analysis algorith…”
MethodRobust medical image compression based on wavelet transform and vector quantization
Ammah, P. N. T., & Owusu, E. (2019) · Informatics in Medicine Unlocked · doi:10.1016/j.imu.2019.100183 · Cited by 74
ProxIQA: A Proxy Approach to Perceptual Optimization of Learned Image Compression
Chen, L.-H., Bampis, C. G., Li, Z., Norkin, A., & Bovik, A. C. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3036752 · Cited by 79
Self-citationExtreme Low-Light Image Enhancement for Surveillance Cameras Using Attention U-Net
Ai, S., & Kwon, J.-W. (2020) · Sensors · doi:10.3390/s20020495 · Cited by 70
“For a fair qualitative comparison, we used the three main attributes that are the most favored in image-quality metrics: image quality assessment [42] for peak signal-to-noise ratio (PSNR), the structural similarity metric (SSIM), and multiscale structural similarity for image quality assessment (MS-SSIM) [43] as show…”
MethodShajkofci, A., & Liebling, M. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.2986880 · Cited by 79
“We computed the scores by taking the difference between the “ground truth" SNR and SSIM of images deconvolved using the PSFs actually used to degrade the images, and the deconvolution results using PSFs regressed with the CNN or other BD techniques.”
MethodInfluentialToward Top-Down Just Noticeable Difference Estimation of Natural Images
Jiang, Q., Liu, Z., Wang, S., Shao, F., & Lin, W. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3174398 · Cited by 83
“Then, the performance of a JND model can be evaluated by measuring the MS-SSIM [41] score and the mean opinion”
MethodLearning Conditional Knowledge Distillation for Degraded-Reference Image Quality Assessment
Zheng, H., Yang, H., Fu, J., Zha, Z.-J., & Luo, J. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01008 · Cited by 94
“, calculating the distance between restored and reference images and mapping such distance to quality scores [21, 36, 37, 44].”
BackgroundDeep Learning for Downscaling Remote Sensing Images: Fusion and super-resolution
Sdraka, M., Papoutsis, I., Psomas, B., Vlachos, K., Ioannidis, K., Karantzalos, K. G., Gialampoukidis, I., & Vrochidis, S. (2022) · IEEE Geoscience and Remote Sensing Magazine · doi:10.1109/mgrs.2022.3171836 · Cited by 79
“METRIC RANGE DESCRIPTION CATEGORY Mean square error (MSE) [ , )0 3 Pixel-based mean square error FR Root-mean-square error (RMSE) [ , )0 3 Pixel-based root-mean-square error FR Mean absolute error (MAE) [ , )0 3 Pixel-based mean absolute error FR Correlation coefficient (CC) [–1, 1] Pixel-based correlation FR Coeffici…”
MethodBASICS: Broad Quality Assessment of Static Point Clouds in a Compression Scenario
Ak, A., Zerman, E., Quach, M., Chetouani, A., Smolić, A., Valenzise, G., & Callet, P. L. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3355642 · Cited by 63
“Bit-per-point is used for the rate, which is shown on the x-axes of the plots, and mean opinion scores are shown on the y-axes of the plots.”
BackgroundMethodLaMoSC: Large Language Model-Driven Semantic Communication System for Visual Transmission
Zhao, Y., Yue, Y., Hou, S., Cheng, B. H., & Huang, Y. (2024) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2024.3401712 · Cited by 84
Low-Complexity Video Quality Assessment Using Temporal Quality Variations
Narwaria, M., Lin, W., & Liu, A. (2012) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2012.2190589 · Cited by 64
Single-Scale Fusion: An Effective Approach to Merging Images
Ancuti, C. O., Ancuți, C., Vleeschouwer, C. D., & Bovik, A. C. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2621674 · Cited by 63
“Besides the TMQI indexes, the last column of the Table shows the SSIM values between MSF and SSF results.”
MethodInfluentialSelf-citationEffective and Efficient Photo-Based PM2.5 Concentration Estimation
Yue, G., Gu, K., & Qiao, J. (2019) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2018.2886091 · Cited by 72
“ity between one high-quality image and its corrupted version can effectively avoid the content’s impact [33]–[35].”
BackgroundReconstruction of natural visual scenes from neural spikes with deep neural networks
Zhang, Y., Jia, S., Zheng, Y., Yu, Z., Tian, Y., Ma, S., Huang, T., & Liu, J. K. (2020) · Neural Networks · doi:10.1016/j.neunet.2020.01.033 · Cited by 59
“Therefore, these measures are a general guideline rather than the precise characteristic, as it seems that there is no perfect measure [42, 43].”
BackgroundA large-scale video codec comparison of x264, x265 and libvpx for practical VOD applications
Cock, J. D., Mavlankar, A., Moorthy, A. K., & Aaron, A. (2016) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2238495 · Cited by 63
Learning to Super Resolve Intensity Images From Events
Mostafavi, M., Choi, J., & Yoon, K. (2020) · 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/cvpr42600.2020.00284 · Cited by 73
“For the quantitative analyses, we use PSNR in dB (logarithmic scale), the structural similarity [25] (SSIM) as a fraction between zero (less similar) to one (fully similar), the mean squared error (MSE), and the perceptual similarity (LPIPS) as a metric to evaluate the similarity of the high level features in two imag…”
MethodChest X-ray Bone Suppression for Improving Classification of Tuberculosis-Consistent Findings
Rajaraman, S., Zamzmi, G., Folio, L. R., Alderson, P. O., & Antani, S. (2021) · Diagnostics · doi:10.3390/diagnostics11050840 · Cited by 68
Reconstructing high fidelity digital rock images using deep convolutional neural networks
Bizhani, M., Ardakani, O. H., & Little, E. C. (2022) · Scientific Reports · doi:10.1038/s41598-022-08170-8 · Cited by 70
Hitchhiker's Guide to Super-Resolution: Introduction and Recent Advances
Moser, B. B., Raue, F., Frolov, S., Palacio, S., Hees, J., & Dengel, A. R. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3243794 · Cited by 86
“The MS-Structural Similarity (MS-SSIM) index [29] is a direct extension of SSIM that incorporates MS.”
MethodVisual Orientation Selectivity Based Structure Description
Wu, J., Lin, W., Shi, G., Zhang, Y., Dong, W., & Chen, Z. (2015) · IEEE Transactions on Image Processing · doi:10.1109/tip.2015.2460467 · Cited by 63
Existing and emerging image quality metrics
Dosselmann, R., & Yang, X. (2006) · Conference proceedings - Canadian Conference on Electrical and Computer Engineering · doi:10.1109/ccece.2005.1557355 · Cited by 79
A Continuous QoE Evaluation Framework for Video Streaming Over HTTP
Eswara, N., Manasa, K., Kommineni, A., Chakraborty, S., Sethuram, H. P., Kuchi, K., Kumar, A., & Channappayya, S. S. (2017) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2017.2742601 · Cited by 62
“Traditional Image Quality Assessment (IQA) metrics used for VQA such as Peak Signal-to-Noise Ratio (PSNR) and Multi Scale - Structural Similarity Index Measure (MS-SSIM) [5] fail to capture temporal distortions present in the video.”
MethodInfluentialSeshadrinathan, K., Pappas, T. N., Safranek, R. J., Chen, J., Wang, Z., Sheikh, H. R., & Bovik, A. C. (2009) · Elsevier eBooks · doi:10.1016/b978-0-12-374457-9.00021-4 · Cited by 76
“Improved versions of the SSIM index that use a frequency decomposition have been proposed [16,17] and our discussion of the relation between the SSIM index and contrast masking models helps us understand the reasons for the improved performance of these metrics.”
MethodSelf-citationAssessing Image Quality Issues for Real-World Problems
Chiu, T.-Y., Zhao, Y., & Gurari, D. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00370 · Cited by 63
“ave access to a high-quality version of each novel image, whether partially or completely. For example, distorted images are evaluated against original, intact images for fullreference IQA algorithms [48,50,57,41,25,6,39] and distorted images are evaluated against partial information about the original, intact images…”
MethodPerceptual Quality Assessment of Omnidirectional Images as Moving Camera Videos
Sui, X., Ma, K., Yao, Y., & Fang, Y. (2021) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2021.3050888 · Cited by 75
“Over the years, SSIM and its multi-scale extension [34] have been regarded as standard “perceptual” metrics to guide the optimization of methods for image denoising [35], image compression [36], image synthesis [37], and video coding [38].”
BackgroundUntrained networks for compressive lensless photography
Monakhova, K., Tran, V., Kuo, G., & Waller, L. (2021) · Optics Express · doi:10.1364/oe.424075 · Cited by 68
“Unlike in the noise-free simulation, our method consistently performs better than FISTA for all amounts of erasures on the MSE, LPIPS Alex, and MS-SSIM metric, resulting in a clearer and sharper reconstruction.”
MethodInfluentialIV-PSNR—The Objective Quality Metric for Immersive Video Applications
Dziembowski, A., Mieloch, D., Stankowski, J., & Grzelka, A. (2022) · IEEE transactions on circuits and systems for video technology (Print) · doi:10.1109/tcsvt.2022.3179575 · Cited by 54
“…Y [13], and WS-PSNR YUV (weighted in the same way as PSNR YUV ) ; CS-PSNR [19]; VMAF [37]; SSIM [23] and a multiscale version of SSIM – MS-SSIM [25]; a pixel-based version of VIF (VIF-P) [32]; PSNR-HVS [16]; PSNR-HVS-M [17]; SFF [33]; PSNR-HA and PSNR-HMA [18]; VSNR [21]; WSNR [22]; 4 metrics…”
BackgroundMethodSTFNet: Self-Supervised Transformer for Infrared and Visible Image Fusion
Liu, Q., Pi, J., Li, X., Yuan, D., He, Z., & Chang, X. (2024) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2024.3352490 · Cited by 73
“The structure similarity loss can be described as: where MSSSIM ( , ) denotes the multi-scale structure similarity [64], and w 1 , w 2 represent the balance parameters.”
BackgroundToken Fusion: Bridging the Gap between Token Pruning and Token Merging
Kim, M., Gao, S., Hsu, Y.-C., Shen, Y., & Jin, H. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00141 · Cited by 206
A fusion-based video quality assessment (fvqa) index
Lin, J. Y., Liu, T.-J., Wu, E. C.-H., & Kuo, C.-C. J. (2014) · Signal and Information Processing Association Annual Summit and Conference (APSIPA), 2014 Asia-Pacific · doi:10.1109/apsipa.2014.7041705 · Cited by 62
Vision-based, real-time retinal image quality assessment
Davis, H., Russell, S. R., Barriga, E. S., et al. (2009) · IEEE International Symposium on Computer-Based Medical Systems (CBMS) · doi:10.1109/cbms.2009.5255437 · Cited by 62
Evaluation of Spatial Gamut Mapping Algorithms
Bonnier, N., Schmitt, F., Brettel, H., et al. (2006) · Color and Imaging Conference · doi:10.2352/cic.2006.14.1.art00011 · Cited by 62
“In this section, we measure the difference between original and gamut-mapped images with four IQMs, CIELab�Eab, S-Cielab�Eab [15], iCAM [16], and an extension to color images of SSIM [ 22 ].”
MethodContinuous Prediction of Streaming Video QoE Using Dynamic Networks
Bampis, C. G., Li, Z., & Bovik, A. C. (2017) · IEEE Signal Processing Letters · doi:10.1109/lsp.2017.2705423 · Cited by 65
Self-citationXu, L., Zeng, X., Huang, Z., Li, W., & Zhang, H. (2019) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2019.101600 · Cited by 68
Belif: Blind Quality Evaluator Of Light Field Image With Tensor Structure Variation Index
Shi, L., Zhao, S., & Chen, Z. (2019) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2019.8803559 · Cited by 78
“In order to demonstrate the effectiveness of BELIF, we compare several FR and NR IQA metrics, including seven 2D-FR metrics [7, 22, 23, 24, 25, 26], three 2D-NR metrics [13, 14, 27], one 3D-FR metric [8], two 3D-NR metrics [15, 18], five multi-view FR metrics [28, 29, 9, 30] and one multi-view NR metric [31].”
MethodCURL: Neural Curve Layers for Global Image Enhancement
Moran, S., McDonagh, S., & Slabaugh, G. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412677 · Cited by 92
“In particular, the multi-scale structural similarity (MSSSIM) [18] between the luminance channels of the ground truth and predicted images enforces a reproduction of the contrast, luminance and structure of the target image [8].”
BackgroundSeo, S., Ki, S., & Kim, M. (2020) · IEEE transactions on circuits and systems for video technology (Print) · doi:10.1109/tcsvt.2020.3030895 · Cited by 63
“780 MS-SSIM [14] 0.”
MethodInfluentialMCL-3D: a database for stereoscopic image quality assessment using 2D-image-plus-depth sourcePreprint
Song, R., Ko, H., & Kuo, C. ‐. J. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1405.1403 · Cited by 86
A Hitchhiker’s Guide to Structural Similarity
Venkataramanan, A. K., Wu, C., Bovik, A. C., Katsavounidis, I. X., & Shahid, Z. (2021) · IEEE Access · doi:10.1109/access.2021.3056504 · Cited by 81
“The SSIM concept reached a higher performance in the form of Multi-Scale SSIM (MS-SSIM) [20], which applies SSIM at five spatial resolutions obtained by successive dyadic sampling.”
MethodSelf-citationReduced-Reference Video Quality Assessment of Compressed Video Sequences
Ma, L., Li, S., & Ngan, K. N. (2012) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2012.2202049 · Cited by 61
Gradient information-based image quality metric
Kim, D.-O., Han, H.-S., & Park, R. (2010) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2010.5506022 · Cited by 62
Subjective and Objective Visual Quality Assessment of Textured 3D Meshes
Guo, J., Vidal, V., Cheng, I., Basu, A., Baskurt, A. M., & Lavoué, G. (2016) · ACM Transactions on Applied Perception · doi:10.1145/2996296 · Cited by 81
Wu, Q., Li, H., Ngan, K. N., & Ma, K. (2017) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2017.2710419 · Cited by 65
“In Table IV, we show the detailed MS-SSIM scores for all test images, where the best results under each QP setting have been highlighted by boldface.”
MethodRobust Image Stitching with Multiple Registrations
Herrmann, C., Wang, C., Bowen, R. S., Keyder, E., Krainin, M., Liu, C., & Ramin Zabih, A. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01216-8_4 · Cited by 63
“We then calculate MS-SSIM [24] or PSNR.”
Disentangling Latent Space for VAE by Label Relevant/Irrelevant Dimensions
Zheng, Z., & Sun, L. (2019) · 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/cvpr.2019.01247 · Cited by 66
“We adopt the metric proposed in [5], which measures the average negative MSSSIM [40] between all pairs in the generated image set X .”
MethodVideo Multimethod Assessment Fusion (VMAF) on 360VR Contents
Orduna, M., Díaz, C., Muñoz, L., Pérez, P., Benito, I., & Garcia, N. N. (2020) · Hispana · Cited by 61
A Study of the Perceptually Weighted Peak Signal-To-Noise Ratio (WPSNR) for Image Compression
Erfurt, J., Helmrich, C. R., Bosse, S., Schwarz, H., Marpe, D., & Wiegand, T. (2019) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2019.8803307 · Cited by 70
“Two IQAs which try to overcome this problem and influenced many researchers in image processing are the structural similarity measure (SSIM) and the multiscale SSIM (MS-SSIM) [2],[3], which assume that the human visual system is highly adapted to extract structural information and compares luminance, contrast and stru…”
Hair Segmentation and Removal in Dermoscopic Images Using Deep Learning
Talavera-Martínez, L., Bibiloni, P., & González-Hidalgo, M. (2020) · IEEE Access · doi:10.1109/access.2020.3047258 · Cited by 71
“Within the second group we consider the Multi-Scale Structural Similarity Index (MSSSIM) [40] and the Universal Quality Image Index (UQI) [41], which measure statistical features locally and then combine them.”
MethodReduced-Complexity End-to-End Variational Autoencoder for on Board Satellite Image Compression
de Oliveira, V. A., Chabert, M., Oberlin, T., Poulliat, C., Bruno, M., Latry, C., Carlavan, M., Henrot, S., Falzon, F., & Camarero, R. (2021) · Remote Sensing · doi:10.3390/rs13030447 · Cited by 80
“This experiment, performed on the proposed architecture integrating the simplified entropy model Ballé(2018)-s-laplacian-N64-M , allows quantifying the impact of M on the performance in terms of both MSE and MS-SSIM for increasing values of the target rate.”
MethodInfluentialSingle Image Reflection Removal with Absorption Effect
Zheng, Q., Shi, B., Chen, J., Jiang, X., Duan, L., & Kot, A. C. (2021) · 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/cvpr46437.2021.01319 · Cited by 82
“Comparisons of quantitative results in terms of SSIM, IS, and PSNR on SIR2-THICK [31], ZC20-ORIEN [50], LY20-DATA [26], SIR2 [31], and ZN18-DATA [21] for reflection removal.”
BackgroundMethodInfluentialVersatile Video Coding Standard: A Review From Coding Tools to Consumers Deployment
Hamidouche, W., Biatek, T., Abdoli, M., et al. (2022) · IEEE Consumer Electronics Magazine · doi:10.1109/mce.2022.3144545 · Cited by 60
“The Bjłntegaard delta bit-rate (BD-rate) metric(20) is used as estimation of the bit-rate variations, using as objective quality metrics PSNR, VMAF(21), and MS-SSIM.(22) For the PSNR metric, the BD-rate variations are computed for the PSNR of each color component (Y, U, and V), then a weighted BD-rate value is compute…”
MethodNTIRE 2021 Challenge on Perceptual Image Quality Assessment
Gu, J., Cai, H., Dong, C., Ren, J., Qiao, Y., Gu, S., Timofte, R., Cheon, M., Yoon, S., Kang, B., et al. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00077 · Cited by 67
“After that, various FR-IQA methods have been proposed to bridge the gap between results of IQA methods and human judgements [55, 57, 59, 44, 58].”
MethodSelf-citationOn the Efficiency of Image Metrics for Evaluating the Visual Quality of 3D Models
Lavoué, G., Larabi, M., & Váša, L. (2015) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2015.2480079 · Cited by 72
Zhang, W., & Liu, H. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2681424 · Cited by 69
“…peak signal-to-noise ratio (PSNR) [1], universal quality index (UQI) [67], structural similarity index (SSIM) [12], multi-scale SSIM (MS-SSIM) [68], visual information fidelity (VIF) [69] and feature similarity index (FSIM) [70]; and four no-reference (NR) OQMs, generalized block-edge…”
BackgroundSemi-Reference Sonar Image Quality Assessment Based on Task and Visual Perception
Chen, W., Gu, K., Zhao, T., Jiang, G., & Callet, P. L. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.2991546 · Cited by 58
“Given that the optimal scale for object recognition depends on the viewing conditions [38], the global structure is described using multiscale geometric analysis in this paper.”
MethodDynamic range independent image quality assessment
Aydın, T. O., Mantiuk, R., Myszkowski, K., et al. (2008) · ACM SIGGRAPH Conference Papers (SIGGRAPH) · doi:10.1145/1399504.1360668 · Cited by 60
“As the authors admit [Wang et al. 2003], a challenging problem is to calibrate its parameters, which are quite “abstract” and thus difficult to derive from simple-stimulus subjective experiments as it is typically performed for contrast-based metrics.”
BackgroundA Probabilistic Quality Representation Approach to Deep Blind Image Quality PredictionPreprint
Zeng, H., Zhang, L., & Bovik, A. C. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1708.08190 · Cited by 97
“The use of full-reference IQA predictions computed by engines like FSIM [5], MS-SSIM [3], or VIF [4] as proxy patch scores is also fraught, since these models are also imperfect.”
MethodSelf-citationvan Dael, M., Verboven, P., Zanella, A., Sijbers, J., & Nicolaï, B. M. (2018) · Postharvest Biology and Technology · doi:10.1016/j.postharvbio.2018.05.020 · Cited by 65
Dynamic Receptive Field Generation for Full-Reference Image Quality Assessment
Kim, W., Nguyen, D. C., Lee, S., & Bovik, A. C. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.2968283 · Cited by 68
“We compared DRF-IQA against ten FR-IQA models: PSNR, SSIM [3], MS-SSIM [44], VIF [9], GMSD [45], FSIMc [14], DoG-SSIMc [25], IFC [46], DeepQA [30], WaDIQaM-FR [27] and also six NR-IQA BLIINDS II [11], BRISQUE [47], BIECON [48], DIQA [31], NIMA [49], and DIQaM-NR [27].”
MethodSelf-citationMaD-DLS: Mean and Deviation of Deep and Local Similarity for Image Quality Assessment
Sim, K., Yang, J., Lu, W. B., & Gao, X. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.3037482 · Cited by 61
NTIRE 2022 Challenge on Perceptual Image Quality Assessment
Gu, J., Cai, H., Dong, C., Ren, J., Timofte, R., Gong, Y., Lao, S., Shi, S., Wang, J., Yang, S., et al. (2022) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw56347.2022.00109 · Cited by 87
“After that, various FR-IQA methods have been proposed to bridge the gap between the results of IQA methods and human judgements [68, 75, 78, 57, 76].”
MethodCui, W., Song, K., Feng, H., Jia, X., Liu, S., & Yan, Y. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3290965 · Cited by 81
“Our loss function can be expressed as where K = 5 denotes the total number of side-output layers, ℓ ( k ) bce , ℓ ( k ) ssim , ℓ ( k ) iou , and ℓ ( k ) cel denote BCE loss [47], SSIM loss [48], IOU loss [49], and CEL loss [20], respectively.”
BackgroundA General Paradigm with Detail-Preserving Conditional Invertible Network for Image Fusion
Wang, W., Deng, L.-J., Ran, R., & Vivone, G. (2023) · International Journal of Computer Vision · doi:10.1007/s11263-023-01924-5 · Cited by 65
Deep Learning in Virtual Try-On: A Comprehensive Survey
Islam, T., Miron, A., Liu, X., & Li, Y. (2024) · IEEE Access · doi:10.1109/access.2024.3368612 · Cited by 97
“However, researchers could have used more advanced metrics, such as MS-SSIM [185] or KID [12], which could have provided a more accurate depiction of their model’s performance.”
MethodD²-JSCC: Digital Deep Joint Source-Channel Coding for Semantic Communications
Huang, J., Yuan, K., Huang, C. Z., & Huang, K. (2025) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2025.3531546 · Cited by 95
“Finally, we evaluate the performance of the proposed scheme using the perceptual metric, i.e., the multi-scale structural similarity index (MS-SSIM) [46].”
BVI-HD: A Video Quality Database for HEVC Compressed and Texture Synthesized Content
Zhang, F., Moss, F. M., Baddeley, R., & Bull, D. (2018) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2018.2817070 · Cited by 64
VR IQA NET: Deep Virtual Reality Image Quality Assessment Using Adversarial Learning
Lim, H.-T., Kim, H. G., & Ro, Y. M. (2018) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2018.8461317 · Cited by 59
“D IQA models, which are PSNR, SSIM [5], MS-SSIM [6], and VIFp [7].”
MethodDeep Lossy Plus Residual Coding for Lossless and Near-Lossless Image Compression
Bai, Y., Liu, X., Wang, K., Ji, X., Wu, X., & Gao, W. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3348486 · Cited by 101
“During training, the compression rates are approximated and minimized with the entropy of the quantized latent variables, while the reconstruction distortion is usually minimized with PSNR or MS-SSIM [32], leading to the rate-distortion optimization [1].”
MethodWang, J., Gao, W., & Li, G. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3362369 · Cited by 70
“We compare PCQA methods with different calculation ways to reflect the efficiency characteristics of each method, including SSIM [37], MS-SSIM [63], Graph-SIM [29], PointSSIM [51], 3D-NSS [45], MM-PCQA [8], and our proposed MOD-PCQA.”
MethodImproved Block Truncation Coding Using Optimized Dot Diffusion
Guo, J.-M., & Liu, Y.-F. (2013) · IEEE Transactions on Image Processing · doi:10.1109/tip.2013.2257812 · Cited by 58
Image Quality Assessment Using Similar Scene as Reference
Liang, Y., Wang, J., Wan, X., Gong, Y., & Zheng, N. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-46454-1_1 · Cited by 65
Zhang, Y., Chandler, D. M., & Mou, X. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2851390 · Cited by 57
“The six classical quality metrics were SSIM [10], multiscale structure similarity (MS-SSIM) [25], feature similarity index (FSIM) [26], gradient similarity (GSIM) [27], visual saliency-induced index (VSI) [28], and gradient magnitude similarity deviation (GMSD) [29].”
BackgroundAdversarial Learning of Semantic Relevance in Text to Image Synthesis
Cha, M., Gwon, Y., & Kung, H. T. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33013272 · Cited by 68
Sikka, A., Peri, S., Virk, J. S., Niyaz, U., & Bathula, D. R. (2025) · The Journal of Precision Medicine: Health and Disease · doi:10.1016/j.premed.2025.100004 · Cited by 58
“Hence, [35] proposed a multi-scale variant known as multi-scale structural similarity index (MSSSIM) that computes SSIM at multiple scales.”
Objective and Subjective QoE Evaluation for Adaptive Point Cloud Streaming
van der Hooft, J., Vega, M. T., Timmerer, C., et al. (2020) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex48832.2020.9123081 · Cited by 57
“Based on the state of the art, we prepared a set of 24 impaired volumetric test videos that were analyzed both subjectively, through a single stimulus approach, and objectively, through the full reference metrics PSNR, SSIM and VQM.”
MethodInfluentialWang, J., Liu, H., Jiang, P., Wang, Z., Sui, Q., & Zhang, F. (2021) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2021.3111445 · Cited by 75
“Multiscale structural similarity index (MSSIM) [64] and SSIM were used to quantify the similarities of the local structures between the inversion results and ground truths.”
MethodSelf-citationOASIS: Only Adversarial Supervision for Semantic Image Synthesis
Sushko, V., Schönfeld, E., Zhang, D., Gall, J., Schiele, B., & Khoreva, A. (2022) · International Journal of Computer Vision · doi:10.1007/s11263-022-01673-x · Cited by 217
“To measure the diversity among synthesized samples in the multi-modal image generation regime, we evaluate MS-SSIM (Wang et al., 2003) and LPIPS (Zhang et al., 2018b) between the images generated from the same label map.”
MethodInfluentialObjective Video Quality Assessment Based on Perceptually Weighted Mean Squared Error
Hu, S., Jin, L., Wang, H., Zhang, Y., Kwong, S., & Kuo, C. ‐. J. (2016) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2016.2556499 · Cited by 57
“In this section, we compare the performance of the proposed method with other benchmarks, including: MS-SSIM [52], VIF [53], ST-MAD [54], VQM [55], MOVIE [56].”
MethodWang, T.-H., Zhang, L., Jia, H.-Z., Li, B.-S., & Shu, H. (2016) · Signal processing. Image communication · doi:10.1016/j.image.2016.04.005 · Cited by 61
“Then a lot of SSIM-based metrics have been proposed in the literature [4-6].”
BackgroundMethodInfluentialPerceptual Reduced-Reference Visual Quality Assessment for Contrast Alteration
Liu, M., Gu, K., Zhai, G.-T., Callet, P. L., & Zhang, W.-J. (2017) · IEEE transactions on broadcasting · doi:10.1109/tbc.2016.2597545 · Cited by 56
2D and 3D Image Quality Assessment: A Survey of Metrics and Challenges
Niu, Y., Zhong, Y., Guo, W., Shi, Y., & Chen, P. (2018) · IEEE Access · doi:10.1109/access.2018.2885818 · Cited by 65
“The cyclopeanmultiscale SSIM (CMSSSIM) index [68] applies a 2D MS-SSIM [18] index to the Gabor cyclopean images of the original and distorted images.”
MethodBlind Image Quality Assessment by Learning from Multiple Annotators
Ma, K.-D., Liu, X., Fang, Y., & Simoncelli, E. P. (2019) · International Conference on Information Photonics · doi:10.1109/icip.2019.8803390 · Cited by 66
“To generate training labels, we used six full-reference IQA models: SSIM [1], MS-SSIM [23], VIF [24], MAD [12], VSI [25] and NLPD [26]; and three BIQA models: NIQE [27], ILNIQE [28] and dipIQ [9].”
MethodSelf-citationStereoscopic Omnidirectional Image Quality Assessment Based on Predictive Coding Theory
Chen, Z., Xu, J., Lin, C., & Zhou, W. (2020) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2020.2968182 · Cited by 60
“Firstly, 2D IQA metrics are used for conventional 2D image quality assessment, we choose five commonly used metrics for comparison, namely PSNR, SSIM [11], MSSSIM [12], FSIM [13], and VSI [64].”
BackgroundMethodZhang, Y., & Chandler, D. M. (2015) · IEEE Transactions on Image Processing · doi:10.1109/tip.2015.2456414 · Cited by 55
“In [54], Ko et al. introduced a binocular perception model called the structural distortion parameter (SDP) to control the three components of SSIM to obtain an overall quality index.”
MethodInfluentialShahidi, F. (2021) · IEEE Access · doi:10.1109/access.2021.3057497 · Cited by 67
“Is proposed by Wang et al. [37].”
Deep learning for velocity model building with common-image gather volumes
Geng, Z., Zhao, Z., Shi, Y., Wu, X., Fomel, S. B., & Sen, M. K. (2021) · Geophysical Journal International · doi:10.1093/gji/ggab385 · Cited by 62
“(5) are borrowed from the original paper (Wang et al. 2003 the step learning rate scheduler by reducing learning rate to 90 per cent after every 10 epochs.”
BackgroundMethodMaking Sense of Meaning: A Survey on Metrics for Semantic and Goal-Oriented Communication
Getu, T. M., Kaddoum, G., & Bennis, M. (2023) · IEEE Access · doi:10.1109/access.2023.3269848 · Cited by 91
AEROBLADE: Training-Free Detection of Latent Diffusion Images Using Autoencoder Reconstruction Error
Ricker, J., Lukovnikov, D., & Fischer, A. (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/cvpr52733.2024.00872 · Cited by 234
“They find that LPIPS outperforms specialized distance metrics like SSIM [52] or MS-SSIM [51].”
MethodCharacterizing perceptual artifacts in compressed video streams
Zeng, K., Zhao, T., Rehman, A., & Wang, Z. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2043128 · Cited by 55
Self-citationPiCode: A New Picture-Embedding 2D Barcode
Chen, C., Huang, W., Zhou, B., et al. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2573592 · Cited by 56
“However, the multi-scale structural similarity (MS-SSIM) metric [30] is employed in our experiment to cater for different viewing conditions, such as different resolutions and physical sizes, which are important for barcodes with different capacities.”
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.