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 401–500 of 6,047
Su, C.-C., Cormack, L. K., & Bovik, A. C. (2015) · IEEE Transactions on Image Processing · doi:10.1109/tip.2015.2409558 · Cited by 100
“, PSNR [41], SSIM [42], and MS-SSIM [11], and 2D NR models, e.”
InfluentialSelf-citationNo-reference image quality assessment algorithms: A survey
Kamble, V. M., & Bhurchandi, K. M. (2015) · Optik · doi:10.1016/j.ijleo.2015.02.093 · Cited by 112
A Comprehensive Benchmark for Single Image Compression Artifact Reduction
Liu, J., Liu, D., Yang, W., Xia, S., Zhang, X., & Dai, Y. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3007828 · Cited by 121
“Four full-reference metrics, including PSNR, PSNR-B [59], SSIM [60], MS-SSIM [61], and two non-reference metrics, including NIQE [55], and BRISQUE [62], are used to evaluate the effectiveness of the proposed method.”
MethodSalehi, P., & Chalechale, A. (2020) · Iranian Conference on Machine Vision and Image Processing · doi:10.1109/mvip49855.2020.9116895 · Cited by 130
“The metrics used for comparison are Structural Similarity index (SSIM) [22], Multi-scale Structural Similarity Index (MS-SSIM) [23], Feature Similarity (FSIM) Index [24], Spatial Correlation Coefficient (SCC), Pearson Correlation Coefficient (PCC) [25], Mean Squared Error (MSE), Root Mean Square Error (RMSE) [26], Pea…”
MethodFuture Video Synthesis With Object Motion Prediction
Wu, Y., Gao, R., Park, J., & Chen, Q. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00558 · Cited by 93
“And the improvement of MS-SSIM is 5%, 7%, 10%, respectively.”
MethodInfluentialAdversarial Objectness Gradient Attacks in Real-time Object Detection Systems
Chow, K.-H., Liu, L., Loper, M. L., Bae, J., Gürsoy, M. E., Truex, S., Wei, W., & Wu, Y. (2020) · IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA) · doi:10.1109/tps-isa50397.2020.00042 · Cited by 138
“For the distortion cost, we consider L 0 , L 2 , L ∞ distances and the structural similarity (SSIM) [24].”
MethodEnergy compaction-based image compression using convolutional autoencoder
Cheng, Z., Sun, H., Takeuchi, M., & Katto, J. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2019.2938345 · Cited by 97
“To achieve high subjective quality in the final experiments, we used a popular MS-SSIM [31] as the distortion term defined by 1 − MS - SSIM(x , ˆx) .”
MethodGeological Facies modeling based on progressive growing of generative adversarial networks (GANs)
Song, S., Mukerji, T., & Hou, J. (2021) · Computational Geosciences · doi:10.1007/s10596-021-10059-w · Cited by 109
“Many researchers use the multi-scale structural similarity index (MS-SSIM) [38] or directly the structural similarity index (SSIM) to evaluate the generator’s output (e.”
MethodBenchmarking of objective quality metrics for HDR image quality assessment
Hanhart, P., Bernardo, M. V., Pereira, M., Pinheiro, A., & Ebrahimi, T. (2015) · EURASIP Journal on Image and Video Processing · doi:10.1186/s13640-015-0091-4 · Cited by 107
Rematas, K., Ritschel, T., Fritz, M., Gavves, E., & Tuytelaars, T. (2016) · 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.2016.488 · Cited by 110
“The first is plain L2 error between all defined pixels of the reflectance map in RGB and the second the SSIM structural difference [36].”
Group MAD Competition? A New Methodology to Compare Objective Image Quality Models
Ma, K., Wu, Q., Wang, Z., Duanmu, Z., Yong, H., Li, H., & Zhang, L. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.184 · Cited by 93
“Third, the best performance overall is obtained by MS-SSIM [34], which is a multi-scale version of SSIM [32] and significantly improves upon it.”
MethodSelf-citationObjective Quality Assessment of Screen Content Images by Uncertainty Weighting
Fang, Y., Yan, J.-B., Liu, J.-Y., Wang, S.-Q., Li, Q., & Guo, Z.-M. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2669840 · Cited by 83
“In this experiment, we perform the comparison experiments by using SFUW and the following existing visual quality metrics: PSNR, SSIM [23], IWSSIM [51], MSSIM [50], VIF [26], IFC [32], MAD [34], GMSD [33], SPQA [5], SQI [17], EMSQA [53] and GSS [52].”
MethodVisual Quality Assessment for Super-Resolved Images: Database and Method
Zhou, F., Yao, R., Liu, B., & Qiu, G. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2898638 · Cited by 104
Liu, X., Zhang, Y., Hu, S., Kwong, S., Kuo, C. ‐. J., & Peng, Q. (2015) · IEEE Transactions on Image Processing · doi:10.1109/tip.2015.2469140 · Cited by 95
“multi-scale structural similarity between images [61].”
Orientation selectivity based visual pattern for reduced-reference image quality assessment
Wu, J., Lin, W., Shi, G., Li, L., & Fang, Y. (2016) · Information Sciences · doi:10.1016/j.ins.2016.02.043 · Cited by 99
“Keywords: Orientation Selectivity Mechanism, Image Quality Assessment, Reduced-Reference, Excitatory and Inhibitory Interactions, Preferred Orientation, Visual Pattern Preprint submitted to Information Sciences February 15, 2016”
BackgroundLeveraging 2D Data to Learn Textured 3D Mesh Generation
Henderson, P. D., Tsiminaki, V., & Lampert, C. H. (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/cvpr42600.2020.00752 · Cited by 117
“Second, we replace the Gaussian log-likelihood log P ( x | z ) by the multi-scale structural similarity (MS-SSIM) metric [50], which (i) allows gradients to propagate across longer distances in the image, and (ii) tends to preserve fine details better.”
MethodDeep Learning for Ultrasound Image Formation: CUBDL Evaluation Framework and Open Datasets
Hyun, D., Wiacek, A., Goudarzi, S., Rothlübbers, S., Asif, A. A., Eickel, K., Eldar, Y. C., Huang, J., Mischi, M., Rivaz, H. J., et al. (2021) · IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control · doi:10.1109/tuffc.2021.3094849 · Cited by 105
Binocular spatial activity and reverse saliency driven no-reference stereopair quality assessment
Liu, L., Liu, B., Su, C.-C., Huang, H., & Bovik, A. C. (2017) · Signal Processing Image Communication · doi:10.1016/j.image.2017.08.011 · Cited by 101
“Moreover, multiscale processing has been shown to improve the efficacy of leading image quality models [6,38].”
Self-citationLearning Image and Video Compression Through Spatial-Temporal Energy Compaction
Cheng, Z., Sun, H., Takeuchi, M., & Katto, J. (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.2019.01031 · Cited by 102
“Measurements To achieve high subjective quality, we used a popular MS-SSIM [30] as distortion term defined by D = 1 − MS-SSIM(x, x̂).”
BackgroundMethodImage Defogging Quality Assessment: Real-World Database and Method
Liu, W., Zhou, F., Lü, T., Duan, J., & Qiu, G. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3033402 · Cited by 96
“In order to improve SSIM, many IQA methods have been designed to evaluate distorted images, such as multi-scale SSIM (MS-SSIM) [54], information content weighted SSIM (IW-SSIM) [53], information fidelity criterion (IFC) [45], visual information fidelity (VIF) [44], etc.”
MethodAn Optical Flow-Based Full Reference Video Quality Assessment Algorithm
Manasa, K., & Channappayya, S. S. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2548247 · Cited by 93
Reflectance and Natural Illumination from Single-Material Specular Objects Using Deep Learning
Georgoulis, S., Rematas, K., Ritschel, T., Gavves, E., Fritz, M., Gool, L. V., & Tuytelaars, T. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2742999 · Cited by 97
“The first is the traditional L2 error between all defined pixels of the reflectance map in RGB and the second is the DSSIM structural difference [55] that excels in measuring the similarity between two images.”
BackgroundMethodAn efficient krill herd algorithm for color image multilevel thresholding segmentation problem
He, L., & Huang, S. (2020) · Applied Soft Computing · doi:10.1016/j.asoc.2020.106063 · Cited by 93
No-Reference Video Quality Assessment Using Natural Spatiotemporal Scene Statistics
Dendi, S. V. R., & Channappayya, S. S. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.2984879 · Cited by 102
“We compared the proposed approach with FRIQA methods like SSIM [5] and MS-SSIM [6], NRIQA methods like BRISQUE [11] and NIQE [8], FRVQA methods like MOVIE [14] and VQM [16], and NRVQA methods like V-BLIINDS [29], VIIDEO [44] and TLVQM [34].”
MethodA Deep Learning Enhanced Novel Software Tool for Laryngeal Dynamics Analysis
Kist, A. M., Gómez, P., Dubrovskiy, D., Schlegel, P., Kunduk, M., Echternach, M., Patel, R. R., Semmler, M., Bohr, C., Dürr, S., et al. (2021) · Journal of Speech Language and Hearing Research · doi:10.1044/2021_jslhr-20-00498 · Cited by 102
MLFF-GAN: A Multilevel Feature Fusion With GAN for Spatiotemporal Remote Sensing Images
Song, B., Liu, P., Li, J., Wang, L., Zhang, L., He, G., Chen, L., & Liu, J. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3169916 · Cited by 114
“In order to obtain the spectral and structural authenticity of the generated remote sensing image, the cosine similarity and the multiscale structural similarity (MS-SSIM) [53] between the real image and the generated image are calculated respectively.”
NeRF-DS: Neural Radiance Fields for Dynamic Specular Objects
Yan, Z., Li, C., & Lee, G. H. (2023) · 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/cvpr52729.2023.00801 · Cited by 123
“The generated images are compared with the ground truth test view images to calculate the following quantitative metrics: MS-SSIM [52], PSNR, LPIPS [58] as in previous works [22, 33, 45].”
MethodInfluentialLatency-Aware Generative Semantic Communications With Pre-Trained Diffusion Models
Qiao, L., Mashhadi, M. B., Gao, Z., Foh, C. H., Xiao, P., & Bennis, M. (2024) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2024.3429295 · Cited by 137
“The quality requirements can represent any semantic distortion or perception [8], [9] requirement of the signal synthesized at the receiver, and can be enforced by various metrics, e.g., MS-SSIM [20], LPIPS [21], FID [22], CLIP [23], etc.”
MethodColor-Image Quality Assessment: From Prediction to Optimization
Preiss, J., Fernandes, F., & Urban, P. (2014) · IEEE Transactions on Image Processing · doi:10.1109/tip.2014.2302684 · Cited by 100
Modeling the Time—Varying Subjective Quality of HTTP Video Streams With Rate Adaptations
Chen, C., Choi, L. K., de Veciana, G., Caramanis, C., Heath, R. W., & Bovik, A. C. (2014) · IEEE Transactions on Image Processing · doi:10.1109/tip.2014.2312613 · Cited by 93
“Since the proposed model predicts the average TVSQ per second, the prediction timescales are suitable for HTTP-based streaming.”
MethodSelf-citationA ParaBoost Method to Image Quality Assessment
Liu, T.-J., Liu, K.-H., Lin, J., Lin, W., & Kuo, C.-C. J. (2017) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2015.2500268 · Cited by 85
Neural Video Coding Using Multiscale Motion Compensation and Spatiotemporal Context Model
Liu, H., Lu, M., Ma, Z., Wang, F., Xie, Z., Cao, X., & Wang, Y. (2020) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3035680 · Cited by 121
“The NVC demonstrated the leading performance with consistent gains across all popular test sequences for both PSNR (Peak signal-to-noise ratio) and MS-SSIM (multiscale structural similarity) [6] distortion metrics.”
BackgroundFast Task-Specific Region Merging for SAR Image Segmentation
Ma, F., Zhang, F., Xiang, D., Yin, Q., & Zhou, Y. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3141125 · Cited by 102
Cheng, J. C. P., & Ma, J. (2015) · Building and Environment · doi:10.1016/j.buildenv.2015.07.019 · Cited by 91
DewarpNet: Single-Image Document Unwarping With Stacked 3D and 2D Regression Networks
Das, S., Ma, K., Shu, Z., Samaras, D., & Shilkrot, R. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00022 · Cited by 95
“We use two image similarity metrics: Multi-Scale Structural Similarity (MS-SSIM) [42] and Local Distortion (LD) [45], as quantitative evaluation criteria, following [23].”
MethodNTIRE 2020 Challenge on Real Image Denoising: Dataset, Methods and Results
Abdelhamed, A., Afifi, M., Timofte, R., Brown, M. S., Cao, Y., Zhang, Z., Zuo, W., Zhang, X., Liu, J., Chen, W., et al. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00256 · Cited by 95
“loss between L1, L2, multi-scale structural similarity (MSSSIM) [41], and/or Laplace gradients.”
BackgroundAdaptive frequency median filter for the salt and pepper denoising problem
Erkan, U., Enginoğlu, S., Thanh, D. N. H., & Hiếu, L. M. (2019) · IET Image Processing · doi:10.1049/iet-ipr.2019.0398 · Cited by 91
A Survey on Multimedia Services QoE Assessment and Machine Learning-Based Prediction
Kougioumtzidis, G., Poulkov, V. K., Zaharis, Z. D., & Lazaridis, P. I. (2022) · IEEE Access · doi:10.1109/access.2022.3149592 · Cited by 114
“It considers the image signal’s sampling density, the proximity among the viewer and the image, and the perceptual capabilities of the viewer’s HVS [56].”
BackgroundUnifying Motion Deblurring and Frame Interpolation with Events
Zhang, X., & Yu, L. (2022) · 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/cvpr52688.2022.01724 · Cited by 126
“We compare to the state-of-the-art frame-based deblurring approach LEVS [11] and event-based methods including EDI [22], LEDVDI [14], eSL-Net [31], RED [34], and evaluate the results by metrics PSNR, SSIM [32] and LPIPS [36].”
MethodInfluentialAn efficient color image quality metric with local-tuned-global model
Gu, K., Zhai, G.-T., Yang, X.-K., & Zhang, W.-J. (2014) · International Conference on Information Photonics · doi:10.1109/icip.2014.7025101 · Cited by 85
Perceptual Quality Assessment of 3d Point Clouds
Su, H., Duanmu, Z., Liu, W., Liu, Q., & Wang, Z. (2019) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2019.8803298 · Cited by 109
“By contrast, projection-based models perform better, among which visual information fidelity in pixel domain (VIFP) [28] achieves the best performance compared to PSNR, structural similarity index (SSIM) [29] and multi-scale structural similarity (MSSSIM) [30].”
MethodSelf-citationAdversarial Monte Carlo denoising with conditioned auxiliary feature modulation
Xu, B., Zhang, J., Wang, R., Xu, K., Yang, Y., Li, C., & Tang, R. (2019) · ACM Transactions on Graphics · doi:10.1145/3355089.3356547 · Cited by 105
“Selecting L2 as loss function will maximize PSNR (peak signal-to-noise ratio) value, but this is not enough to guarantee perceptual quality [Wang et al. 2003].”
BackgroundHuang, L., Liu, T., Yang, X., Luo, Y., Rivenson, Y., & Özcan, A. (2021) · ACS Photonics · doi:10.1021/acsphotonics.1c00337 · Cited by 99
SwinSTFM: Remote Sensing Spatiotemporal Fusion Using Swin Transformer
Chen, G., Jiao, P., Hu, Q. X., Xiao, L., & Ye, Z. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3182809 · Cited by 102
“Among them, the Charbonnier loss in LapSRN [42] is used for pixel-level image supervision, and the multiscale structural similarity (MS-SSIM) [43] is used to measure the overall similarity between the generated image and the real fine image.”
MethodA Dynamic Multi-Scale Voxel Flow Network for Video Prediction
Hu, X., Huang, Z., Huang, A., Xu, J., & Zhou, S. (2023) · 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/cvpr52729.2023.00593 · Cited by 144
“On MSSSIM and LPIPS, our DMVFN achieves much better results than other methods in both short-term and long-term video prediction.”
MethodPerceptual Evaluation of Light Field Image
Shi, L., Zhao, S., Zhou, W., & Chen, Z. (2018) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2018.8451077 · Cited by 106
“5675 MS-SSIM [22] 0.”
A Study of High Frame Rate Video Formats
Mackin, A., Zhang, F., & Bull, D. (2018) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2018.2880603 · Cited by 97
“The quality metrics considered here are: MS-SSIM [60], PSNR [61], SSIM [62], VIF [63], VSNR [64], ST-MAD [65] and VQM [66].”
MethodPixelated Semantic Colorization
Zhao, J., Han, J., Shao, L., & Snoek, C. G. M. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01271-4 · Cited by 103
“Following Guadarrama et al (2017), we compare two outputs from the same gray-scale image with multiscale structural similarity (SSIM) (Wang rather than recover the ground-truth.”
Fast structured illumination microscopy via deep learning
Ling, C., Zhang, C., Wang, M., Meng, F., Du, L., & Yuan, X. (2020) · Photonics Research · doi:10.1364/prj.396122 · Cited by 89
Rendering Natural Camera Bokeh Effect with Deep Learning
Ignatov, A., Patel, J., & Timofte, R. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00217 · Cited by 114
“For each network, we computed PSRN, SSIM [18] and LPIPS [20] metrics on the test subset of the dataset.”
MethodInfluentialAdvances in Video Compression System Using Deep Neural Network: A Review and Case Studies
Ding, D., Ma, Z., Chen, D., Chen, Q., Liu, Z. S.-J., & Zhu, F. M. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2021.3059994 · Cited by 116
“There has been notable development in both subjective and objective video quality assessments, yielding several well-known metrics, such as SSIM [230], just-noticeable-distortion (JND) [231], and VMAF [232], some of which are actively adopted for the evaluation of video algorithms, application…”
Understanding and simplifying the structural similarity metric
Rouse, D., & Hemami, S. S. (2008) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2008.4711973 · Cited by 96
A Deep Journey into Super-resolution: A surveyPreprint
Anwar, S., Khan, S. H., & Barnes, N. M. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.07523 · Cited by 86
“perceptual quality [103]. To counter this issue, several perceptual loss measures have been proposed in the literature. The conventional perceptual metrics were fixed e.g., SSIM [93], multi-scale SSIM [104], while more recent ones are learned to model human perception of images e.g., LPIPS [105] and PieAPP [106]. Each…”
Gradient Direction for Screen Content Image Quality Assessment
Ni, Z.-K., Ma, L., Zeng, H., Cai, C., & Ma, K.-K. (2016) · IEEE Signal Processing Letters · doi:10.1109/lsp.2016.2599294 · Cited by 80
With Great Training Comes Great Vulnerability: Practical Attacks against Transfer Learning
Wang, B., Yao, Y., Viswanath, B., Zheng, H.-T., & Zhao, B. Y. (2018) · USENIX Security Symposium · Cited by 139
Wang, H., Wang, G., Li, Y., Zhang, D., & Lin, L. (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/cvpr42600.2020.00042 · Cited by 111
“We can see that at the same magnitude, the adversarial example generated under the supervision of MS-SSIM are much better than those generated under the supervision of SSIM and without any supervision.”
MethodInfluentialTowards compound stereo-video quality metric: a specific encoder-based framework
Boev, A., Gotchev, A. P., Egiazarian, K., Aksay, A., & Akar, G. B. (2006) · IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI) · doi:10.1109/ssiai.2006.1633754 · Cited by 88
Gu, K., Qiao, J., Lee, S., Liu, H., Lin, W., & Callet, P. L. (2019) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2019.2906768 · Cited by 90
“Multiscale analysis is an important attribute of the human visual system (HVS), which has been widely used in many image processing fields, for example, quality evaluation [38] and saliency detection [39].”
BackgroundMethodGenerative adversarial networks: a survey on applications and challenges
Kumar, M. R. P., & Jayagopal, P. (2020) · International Journal of Multimedia Information Retrieval · doi:10.1007/s13735-020-00196-w · Cited by 100
“[104] extended this metric to multi-scale to assess the quality of images by integrating different image resolutions.”
BackgroundRetrospective Motion Correction in Multishot MRI using Generative Adversarial Network
Usman, M., Latif, S., Asim, M., Lee, B.-D., & Qadir, J. (2020) · Scientific Reports · doi:10.1038/s41598-020-61705-9 · Cited by 86
Semantic Perceptual Image Compression Using Deep Convolution Networks
Prakash, A., Moran, N., Garber, S., DiLillo, A., & Storer, J. A. (2017) · DCC · doi:10.1109/dcc.2017.56 · Cited by 88
“Richter et al [xxx˙richter2009ms][xxx˙richter2011ssim] addressed structural similarity (SSIM[9] and MS-SSIM[10]) for jpeg and jpeg 2000.”
BackgroundNo-Reference Image Quality Assessment by Wide-Perceptual-Domain Scorer Ensemble Method
Liu, T.-J., & Liu, K.-H. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2771422 · Cited by 90
“A single-scale WPDSE method described in the previous section has the disadvantage of not considering viewing conditions [3] and characteristics of Human Visual System (HVS).”
MethodSingle-Shot Autofocusing of Microscopy Images Using Deep Learning
Luo, Y., Huang, L., Rivenson, Y., & Özcan, A. (2021) · ACS Photonics · doi:10.1021/acsphotonics.0c01774 · Cited by 91
Learning Content-Weighted Deep Image Compression
Li, M., Zuo, W., Gu, S., You, J., & Zhang, D. D. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.2983926 · Cited by 107
Blind quality assessment for image superresolution using deep two-stream convolutional networks
Zhou, W., Jiang, Q., Wang, Y., Chen, Z., & Li, W. (2020) · Information Sciences · doi:10.1016/j.ins.2020.04.030 · Cited by 96
“For the SR quality database [17], we compare the performance of our proposed DeepSRQ method with three classical FR-IQA metrics, namely PSNR, SSIM [33], and MS-SSIM [35].”
MethodInfluentialStructure-Aware Motion Deblurring Using Multi-Adversarial Optimized CycleGAN
Wen, Y., Chen, J., Sheng, B., Chen, Z., Li, P., Tan, P., & Lee, T. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3092814 · Cited by 87
“Combing with the perceptual loss [18] and multi-scale structural similarity (MS-SSIM [19]) loss, we obtain significantly better image deblurring results than most of the existing methods.”
ResultMulti–Scale Synthesized View Assessment Based on Morphological Pyramids
Sandic-Stankovic, D. D., Kukolj, D., & Callet, P. L. (2016) · Journal of Electrical Engineering · doi:10.1515/jee-2016-0001 · Cited by 84
Blind Deep S3D Image Quality Evaluation via Local to Global Feature Aggregation
Oh, H., Ahn, S., Kim, J., & Lee, S. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2725584 · Cited by 87
BARNet: Boundary Aware Refinement Network for Crack Detection
Guo, J.-M., Markoni, H., & Lee, J.-D. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3069135 · Cited by 87
NTIRE 2021 NonHomogeneous Dehazing Challenge Report
Ancuti, C. O., Ancuți, C., Vasluianu, F.-A., Timofte, R., Fu, M., Liu, H., Yu, Y., Chen, J., Wang, K., Chang, J., et al. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00074 · Cited by 131
“The overall loss function is a linear combination of smooth L1 loss Ll1, MS-SSIM loss[60] LSSIM , perceptual loss[35] Lperceptual, and adversarial loss Ladv , as shown in Equation 2.”
MethodRoad Anomaly Detection by Partial Image Reconstruction with Segmentation Coupling
Vojir, T., Sipka, T., Aljundi, R., Chumerin, N., Reino, D. O., & Matas, J. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01536 · Cited by 116
“( iii ) The reconstructed RGB image ˆ I is compared to the input image I using the structural similarity index measure [44], denoted as SSIM, following the Eq.”
MethodFu, H., Liang, F., Lin, J., Li, B., Akbari, M. A., Liang, J., Zhang, G., Liu, D., Tu, C., & Han, J. (2023) · IEEE Transactions on Image Processing · doi:10.1109/tip.2023.3263099 · Cited by 132
“MS-SSIM metrics [4], and shows great potentials.”
Unifying analysis of full reference image quality assessment
Seshadrinathan, K., & Bovik, A. C. (2008) · 2008 15th IEEE International Conference on Image Processing · doi:10.1109/icip.2008.4711976 · Cited by 85
“Our analysis supports these observations and explicitly links SSIM and contrast masking models.”
MethodSelf-citationSingle Image Dehazing Based on Dark Channel Prior and Energy Minimization
Zhu, M., He, B., & Wu, Q. (2017) · IEEE Signal Processing Letters · doi:10.1109/lsp.2017.2780886 · Cited by 87
“Haze removal quality is measured by multiple metrics: MSE, CIEDE2000 [35], SSIM [36], and MS-SSIM [37].”
MethodAn Integrated Platform for Live 3D Human Reconstruction and Motion Capturing
Alexiadis, D. S., Chatzitofis, A., Zioulis, N., Zoidi, O., Louizis, G., Zarpalas, D., & Daras, P. (2016) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2016.2576922 · Cited by 95
“WMS3IM is calculated from WMS3IMk(u) = ∏3j=1 l j j c β j j s γ j j , where j = 1, 2, 3 stands for the scale and the constants α j , β j , γ j have been set based on the psychovisual experiments of [40], and more specifically {α1, α2, α3} = {0, 0, 0.”
MethodModified-BRISQUE as no reference image quality assessment for structural MR images
Chow, L. S., & Rajagopal, H. (2017) · Magnetic Resonance Imaging · doi:10.1016/j.mri.2017.07.016 · Cited by 92
Blind Video Quality Assessment With Weakly Supervised Learning and Resampling Strategy
Zhang, Y., Gao, X., He, L., Lu, W. B., & He, R. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2868063 · Cited by 86
“In this paper, six top-performing FR-I/VQA metrics (PSNR, SSIM [41], MS-SSIM [24], VIF [3], STMAD [6], ViS3 [7]) are included for comparison.”
MethodPfeiffer, M., Funke, I., Robu, M., Bodenstedt, S., Strenger, L., Engelhardt, S., Roß, T., Clarkson, M. J., Gurusamy, K. S., Davidson, B. R., et al. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-32254-0_14 · Cited by 122
“We add a multi-scale structural similarity [12] loss between an image a and its translation GB(a) (and similarly in the other direction).”
BackgroundMethodWu, X., Zhang, Y., Fan, C., Hou, J., & Kwong, S. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3101484 · Cited by 100
“…[10], 11 different 2D IQA submetrics were used to do the quality prediction, which are error based (NQM [56]), structural similarity based (SSIM [51], MS-SSIM [57], IW-SSIM [58], GSM [59], GMSD [60], RF-SIM [61], SR-SIM [62], VSI [63]), and natural scene statistics based (IFC [64], VIF [53]).”
MethodVisual quality assessment: recent developments, coding applications and future trends
Liu, T.-J., Lin, Y.-C., Lin, W., & Kuo, C. ‐. J. (2013) · APSIPA Transactions on Signal and Information Processing · doi:10.1017/atsip.2013.5 · Cited by 93
Kuang, H., Zhang, X., Li, Y.-J., Chan, L. L. H., & Yan, H. (2017) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2016.2598192 · Cited by 84
Recurrent and Dynamic Models for Predicting Streaming Video Quality of Experience
Bampis, C. G., Li, Z., Katsavounidis, I., & Bovik, A. C. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2815842 · Cited by 84
“1) [32], MS-SSIM [76], SSIM [77] and ST-RRED [33] (see Table V).”
Self-citationStreaming video over HTTP with consistent quality
Li, Z., Begen, A. C., Gahm, J., Shan, Y., Osler, B., & Oran, D. (2014) · Proceedings of the 5th ACM Multimedia Systems Conference · doi:10.1145/2557642.2557658 · Cited by 86
“We define the optimization objective to be an alpha-fairness utility function [11, 18] of some quality metric (for example, MSE/PSNR [1], Sarnoff/PQR [4], MS-SSIM [20, 21], VQM [16], STRRED [17]), which is generic enough to cover a range of candidates.”
MethodVariational Autoencoder for Low Bit-rate Image Compression
Zhou, L., Cai, C., Gao, Y., Su, S., & Wu, J. (2018) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · Cited by 151
Group Maximum Differentiation Competition: Model Comparison with Few Samples
Ma, K., Duanmu, Z., Wang, Z., Wu, Q., Liu, W., Yong, H., Li, H., & Zhang, L. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2889948 · Cited by 93
“4 exhibit approximately the same perceptual quality (in agreement with MS-SSIM [33]) and those in the second row have drastically different perceptual quality (in disagreement with SSIM [27]).”
BackgroundMethodInfluentialSelf-citationDeep learning extended depth-of-field microscope for fast and slide-free histology
Jin, L., Tang, Y., Wu, Y., Coole, J. B., Tan, M. T., Zhao, X., Badaoui, H., Robinson, J. T., Williams, M. D., Gillenwater, A. M., et al. (2020) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2013571117 · Cited by 88
Herrmann, J., Koerzdoerfer, G., Nickel, D. M., Mostapha, M., Nadar, M. S., Gassenmaier, S., Kuestner, T., & Othman, A. E. (2021) · Diagnostics · doi:10.3390/diagnostics11081484 · Cited by 90
“work is first trained to minimize the combined L1 and a multi‐scale version of the struc‐ tural similarity (SSIM) content losses between network prediction and ground truth im‐ ages [29].”
Perceptual Quality Assessment of Low-light Image Enhancement
Zhai, G., Sun, W., Min, X., & Zhou, J. (2021) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3457905 · Cited by 92
“Variance similarity has been widely used in many IQA methods to measure the image structure similarity [74, 75], since it is sensitive to structure change.”
MethodA Comprehensive Review of Deep Learning-Based Real-World Image Restoration
Zhai, L., Wang, Y., Cui, S., & Zhou, Y. (2023) · IEEE Access · doi:10.1109/access.2023.3250616 · Cited by 116
“As an FR IQA, multi-scale structural similarity (MS-SSIM) [73] first performs contrast comparison, structure comparison, and luminance comparison on multi-scale images, and then combines the measurement at different scales.”
MethodMulti-objective Whale Optimization Algorithm for Multilevel Thresholding Segmentation
Elaziz, M. A., Ewees, A. A., Hassanien, A. E., Mudhsh, M., & Xiong, S. (2017) · Studies in computational intelligence · doi:10.1007/978-3-319-63754-9_2 · Cited by 86
Sun, W., Gu, K., Ma, S., Zhu, W., Liu, N., & Zhai, G. (2018) · 2018 IEEE 20th International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp.2018.8547102 · Cited by 78
“These IQA models are Peak signalto-noise ratio (PSNR), Structural Similarity (SSIM) [15], Multi-SSIM (MS-SSIM) [16], Information Content Weight SSIM (IW-SSIM) [17], Analysis of Distortion Distribution for Pooling in SSIM (ADD-SSIM) [18], Visual Signal-toNoise ratio (VSNR) [19], Internal Generative Mechanism (IGM) [20]…”
MethodDeep Learning-Based Image and Video Inpainting: A Survey
Quan, W., Chen, J., Liu, Y., Yan, D.-M., & Wonka, P. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-023-01977-6 · Cited by 138
“These measurements are combined with appropriate weights (Wang et al., 2003).”
InfluentialA survey on Deep-Learning-based image steganography
Song, B., Wei, P., Wu, S., Lin, Y., & Zhou, W. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124390 · Cited by 112
Golestaneh, S. A., & Karam, L. J. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2601821 · Cited by 87
“Moreover, the proposed REDLOG index performs nearly as well as the popular MS-SSIM [57].”
MethodNMF-Based Image Quality Assessment Using Extreme Learning Machine
Wang, S., Deng, C., Lin, W., Huang, G.-B., & Zhao, B. (2016) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2015.2512852 · Cited by 81
“weighted multiplier, which is the same as the MS-SSIM [4] and IW-SSIM [5].”
MethodInfluentialState‐of‐art analysis of image denoising methods using convolutional neural networks
Thakur, R. S., Yadav, R. N., & Gupta, L. (2019) · IET Image Processing · doi:10.1049/iet-ipr.2019.0157 · Cited by 87
“…enhancement factor (IEF), universal quality index (QI), average difference (AD), maximum difference (MD), structural content (SC), correlation quality (CQ), image fidelity (IF), Laplacian MSE (LMSE), peak MSE (PMSE), normalised absolute error (NAE), normalised MSE (NMSE), and L P -norm [26–28].”
Background3FabRec: Fast Few-Shot Face Alignment by Reconstruction
Browatzki, B., & Wallraven, C. (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/cvpr42600.2020.00615 · Cited by 98
“Structural image loss: To penalize reconstructions that do not align facial structures well with input images, we add a structural image loss based on the SSIM [48] image similarity metric, which measures contrast c(a, b) and correlation s(a, b) between two image windows a and b:”
MethodInfluentialMetaInv-Net: Meta Inversion Network for Sparse View CT Image Reconstruction
Zhang, H., Liu, B., Yu, H., & Dong, B. (2020) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2020.3033541 · Cited by 94
“reas-Test set. It has a different distribution to the training set. Quantitative results are measured by the multi-scale SSIM index (with level=5, Gaussian kernel size=11, and standard deviation=1.5) [91], [92] and PSNR. D. Quantitative Measure at Different Scanning Settings In this experiment, we evaluate the perform…”
MethodInfluentialADMNet: Attention-Guided Densely Multi-Scale Network for Lightweight Salient Object Detection
Zhou, X., Shen, K., & Liu, Z. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3413529 · Cited by 104
“…[24] for training, and thus the total loss function L can be formulated as follows where G denotes the ground-truth saliency map, Up denotes that the upsampling rate is set to 2 i for S i , and l ssim i , l bce i and l ioui refer to SSIM loss [76], BCE loss [77] and IoU loss [78], respectively.”
Meng, S., Wu, S., Zhang, J., Cheng, J., Zhou, H., & Zhang, Q. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3416309 · Cited by 110
“The multi-scale structural similarity index (MS-SSIM) [90] is an extension of SSIM.”
BackgroundMethod
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.