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 901–1,000 of 6,047
Efficient Video Quality Assessment Along Temporal Trajectories
Moorthy, A. K., & Bovik, A. C. (2010) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2010.2087470 · Cited by 49
Self-citationPEID: A Perceptually Encrypted Image Database for Visual Security Evaluation
Guo, S., Xiang, T., Li, X., & Yang, Y. (2019) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2935415 · Cited by 43
Assessment of QoE for Video and Audio in WebRTC Applications Using Full-Reference Models
García, B., Gortázar, F., Gallego, M., & Hines, A. (2020) · Electronics · doi:10.3390/electronics9030462 · Cited by 54
“There are several models derived from SSIM, for example, MS-SSIM (Multi-Scale SSIM), conducted over multiples scales of subsampling [12].”
MethodImage Compression with Encoder-Decoder Matched Semantic Segmentation
Hoang, T. M., Zhou, J., & Fan, Y. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00088 · Cited by 53
“We use both PSNR and MS-SSIM [15] metrics to evaluate our results.”
MethodLiang, X., Nguyen, D. T. D., & Jiang, S. B. (2020) · Machine Learning Science and Technology · doi:10.1088/2632-2153/abb214 · Cited by 47
Foreground-Background Parallel Compression With Residual Encoding for Surveillance Video
Wu, L., Huang, K., Shen, H., & Gao, L. (2020) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3027741 · Cited by 44
“The experimental results show that our model outperforms the widely used algorithms, such as H.264 and H.265 when measured by the Multiscale Structural Similarity Index (MS-SSIM) [7] and Peak Signal to Noise Ratio (PSNR).”
A high order PDE-constrained optimization for the image denoising problem
Afraites, L., Hadri, A., Laghrib, A., & Nachaoui, M. (2020) · Inverse Problems in Science and Engineering · doi:10.1080/17415977.2020.1867547 · Cited by 43
“In fact, we provide two metrics: PSNR [56] and the mean structure similarity (SSIM) [57].”
MethodMSLPNet: multi-scale location perception network for dental panoramic X-ray image segmentation
Chen, Q. Y., Zhao, Y., Liu, Y., Sun, Y., Yang, C., Li, P., Zhang, L., & Gao, C. (2021) · Neural Computing and Applications · doi:10.1007/s00521-021-05790-5 · Cited by 51
“However, SSIM ignores the diversity information hidden in the image of different scales proposed by MS-SSIM [49].”
BackgroundCombined image similarity index
Okarma, K. (2012) · Optical Review · doi:10.1007/s10043-012-0055-1 · Cited by 37
Stochastic Image Denoising by Sampling from the Posterior Distribution
Kawar, B., Vaksman, G., & Elad, M. (2021) · 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) · doi:10.1109/iccvw54120.2021.00213 · Cited by 42
“Many alternative distance measures to the MSE have been suggested, including SSIM [48], MS-SSIM [49], IFC [38], VIF [37], VSNR [14], and FSIM [55].”
MethodQuality assessment metric of stereo images considering cyclopean integration and visual saliency
Yang, J., Wang, Y., Li, B., Lu, W. B., Meng, Q., Lv, Z., Zhao, D., & Gao, Z. (2016) · Information Sciences · doi:10.1016/j.ins.2016.09.004 · Cited by 41
“The proposed framework significantly boosted the performance of the corresponding SSIM and MS-SSIM, especially the NC model.”
BackgroundMethodInfluentialAutomatic Inspection and Evaluation System for Pavement Distress
Dong, H., Song, K., Wang, Y., Yan, Y., & Jiang, P. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3113802 · Cited by 44
No-Reference Image Quality Assessment by Hallucinating Pristine Features
Chen, B., Zhu, L., Kong, C., Zhu, H., Wang, S., & Li, Z. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3205770 · Cited by 55
“Recently, there has been a demonstrated success for developing the FR quality measures, including the Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM) [5], Multiscale SSIM (MSSSIM) [6], Visual Saliency-Induced Index (VSI) [7], Median Absolute Deviation (MAD) [8] and Visual Information Fidelity (VI…”
MethodSpatial Decomposition and Temporal Fusion Based Inter Prediction for Learned Video Compression
Sheng, X., Li, L., Liu, D., & Li, H. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3360248 · Cited by 74
“We use PSNR and MS-SSIM [73] to measure the quality of the reconstructed frames in comparison to the original frames.”
MethodA Coding Framework and Benchmark Towards Low-Bitrate Video Understanding
Tian, Y., Lu, G. S., Yan, Y., Zhai, G., Chen, L., & Gao, Z. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3367879 · Cited by 65
“We also evaluate the video coding performance of our framework on HEVC Test Sequences [50], using the traditional visual quality metrics, i.e. , Peak Signal-to-Noise Ratio (PSNR) and MS-SSIM [55].”
Method3DTA: No-Reference 3D Point Cloud Quality Assessment With Twin Attention
Zhu, L., Cheng, J., Wang, X., Su, H., Yang, H., Yuan, H., & Korhonen, J. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3407698 · Cited by 49
Transferable Multimodal Attack on Vision-Language Pre-training Models
Wang, H., Dong, K., Zhu, Z., Qin, H., Liu, A., Fang, X., Wang, J., & Liu, X. (2024) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp54263.2024.00102 · Cited by 96
“To achieve this, we suggest incorporating the structure loss based on SSIM [36] into the optimization process, as it directly influences the structural features of images [40]: where µ d v , µ d v ⊕ δ v and σ d v , σ d v ⊕ δ v are the mean and standard deviations of d v and d v ⊕ δ v respectively,…”
MethodA variation-based ring artifact correction method with sparse constraint for flat-detector CT
Yan, L., Wu, T., Zhong, S., et al. (2016) · Physics in Medicine and Biology · doi:10.1088/0031-9155/61/3/1278 · Cited by 41
UNIQUE: Unsupervised Image Quality Estimation
Temel, D., Prabhushankar, M., & AlRegib, G. (2016) · IEEE Signal Processing Letters · doi:10.1109/lsp.2016.2601119 · Cited by 53
Occlusion-aware Video Temporal Consistency
Yao, C.-H., Chang, C.-Y., & Chien, S. (2017) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3123266.3123363 · Cited by 41
“[32], natural visual feature based video quality metric (VQM) of Pinson et al.”
BackgroundZhang, Y., & Chandler, D. M. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2857413 · Cited by 42
“The six FR methods were SSIM [56], multiscale structure similarity (MS-SSIM) [57], feature similarity index (FSIM) [58], visual information fidelity (VIF) [25], most apparent distortion (MAD) [59], and gradient magnitude similarity deviation (GMSD) [60].”
MethodA High-Definition Diversity-Scene Database for Image Quality Assessment
Liu, T.-J., Liu, H., Pei, S., & Liu, K.-H. (2018) · IEEE Access · doi:10.1109/access.2018.2864514 · Cited by 46
“MS-SSIM [37] defeats all other formulabased models on Contrast change, and VIF [41] has the best performance for Brightness change.”
MethodStudy of Subjective Quality and Objective Blind Quality Prediction of Stereoscopic Videos
Appina, B., Dendi, S. V. R., Manasa, K., Channappayya, S. S., & Bovik, A. C. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2914950 · Cited by 43
“SSIM [59], MS-SSIM [60], VIF [61] are 2D FR IQA models, and BRISQUE [62] is a supervised 2D NR IQA model.”
MethodSelf-citationEstimating Depth-Salient Edges and Its Application to Stereoscopic Image Quality Assessment
Khan, S., & Channappayya, S. S. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2860279 · Cited by 41
“The quality score is computed by applying the MS-SSIM [9] index to the reference and test cyclopean images.”
MethodComponent Attention Guided Face Super-Resolution Network: CAGFace
Kalarot, R., Li, T., & Porikli, F. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093399 · Cited by 53
“Multi-Scale Structural Similarity Index Measure (MS-SSIM) [58] adapts to the variations of viewing conditions, and Feature Similarity Index (FSIM) [71] extends SSIM to feature space.”
MethodScreen Content Video Quality Assessment: Subjective and Objective Study
Cheng, S., Zeng, H., Chen, J., Hou, J., Zhu, J., & Ma, K. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3018256 · Cited by 45
“These 16 objective IQA/VQA models include seven classical PSNR, SSIM [13], MSSIM [14], VIF [15], GMSD [16], MOVIE [18], STMAD [20], and nine state-of-the-art PSIM [17], SQMS [36], ESIM [25], SVQI [38], MDOGS [26], SpEED [23], ViS3 [21], VMAF [22], and FAST [45].”
MethodDeep learning in denoising of micro-computed tomography images of rock samples
Sidorenko, M., Orlov, D., Ebadi, M., et al. (2021) · Computers & Geosciences · doi:10.1016/j.cageo.2021.104716 · Cited by 41
Li, A., Guo, J., Yang, H., Salim, F. D., & Chen, Y. (2021) · IEEE/ACM International Conference on Internet-of-Things Design and Implementation (IoTDI) · doi:10.1145/3450268.3453519 · Cited by 78
“The results show that the quality of reconstructed images from the obfuscated features is significantly decreased from 0.9458 to 0.3175 in terms of MS-SSIM, indicating that the person on the image is hardly reidentifiable visually.”
MethodInfluentialPerceptual Learned Source-Channel Coding for High-Fidelity Image Semantic Transmission
Wang, J., Wang, S., Dai, J., Si, Z., Zhou, D., & Niu, K. (2022) · Globecom · doi:10.1109/globecom48099.2022.10001359 · Cited by 58
“3) Metrics: We evaluate our models and the comparison schemes in classical metric PSNR and MS-SSIM [5] as well as the emerging deep learning based perceptual metric LPIPS [7], DISTS [21], FID [22].”
BackgroundMethodInfluentialDeep Image Steganography Using Transformer and Recursive Permutation
Wang, Z., Zhou, M., Liu, B., & Li, T. (2022) · Entropy · doi:10.3390/e24070878 · Cited by 46
“There exist many image quality assessment metrics, such as PSNR, structural similarity (SSIM) [54,55], feature similarity (FSIM) [56], and gradient magnitude similarity deviation (GMSD) [57].”
BackgroundSelf-citationComputationally-Efficient Neural Image Compression with Shallow Decoders
Yang, Y., & Mandt, S. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00055 · Cited by 61
“Additionally, we examine the R-D performance using the more perceptually relevant MS-SSIM metric [48].”
MethodBlind Quality Assessment of Dense 3D Point Clouds with Structure Guided Resampling
Zhou, W., Yang, Q., Chen, W., Jiang, Q., Zhai, G., & Lin, W. (2024) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3664199 · Cited by 70
“For example, the SSIM [52] and its variant multiscale SSIM [54] as well as the visual information fidelity in the pixel domain [41] are often employed to predict the visual quality of converted 2D images.”
An Efficient Federated Learning Framework for Training Semantic Communication Systems
Nguyen, L. X., Le, H. Q., Tun, Y. L., Aung, P. S., Tun, Y. K., Han, Z., & Hong, C. S. (2024) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2024.3401140 · Cited by 64
“In this subsection, we evaluate the efficiency of the proposed method FedLol with FedAvg, FedProx, and MOON benchmarks in the PSNR metric and the MS-SSIM metric [26], which is a popular metric for evaluating the quality of the reconstructed image.”
MethodLearning Hazing to Dehazing: Towards Realistic Haze Generation for Real-World Image Dehazing
Wang, R., Zheng, Y., Zhang, Z., Li, C., Liu, S., Zhai, G., & Liu, X. (2025) · 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/cvpr52734.2025.02150 · Cited by 63
“We then employ the MS-SSIM loss [45], which is sensitive to structural similarity, as our fidelity metric.”
Structure-Preserving Hybrid Digital-Analog Video Delivery in Wireless Networks
He, D., Luo, C., Lan, C., et al. (2015) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2015.2451956 · Cited by 38
“In the future, we plan to extend our evaluation by using other objective measures such as VIF [18], MS-SSIM [21] and FSIM [25], as well as the subjective visual quality assessment tests.”
MethodSimilarity Assessment Model for Chinese Sign Language Videos
Wang, L., Wang, R., Kong, D., & Yin, B. (2014) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2014.2298382 · Cited by 44
“1) Visual Based Video Quality Assessment: Temporal and spatial distribution of pixels [35], [36], [37], HVS (Human Visual System) feature [38], [39] and structure distortion [40], [41], [42] are common visual features in video quality assessment.”
BackgroundHigh dynamic range video reconstruction from a stereo camera setup
Bätz, M., Richter, T., Garbas, J.-U., Papst, A., Seiler, J., & Kaup, A. (2013) · Signal Processing Image Communication · doi:10.1016/j.image.2013.08.016 · Cited by 38
Gu, K., Zhai, G., Yang, X., Zhang, W., & Liang, L. (2013) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2013.6607462 · Cited by 46
“Since then, many improved SSIM-type of IQA metrics have been developed and obtained higher prediction performance as full-reference (FR) IQA metrics, such as MS-SSIM [2], IW-SSIM [3], SNW-SSIM [4], SC-SSIM [5] and DIP [6].”
BackgroundMethodDevelopment of improved SSIM quality index for compressed medical images
Kumar, B., Kumar, S. B., & Kumar, C. (2013) · International Conference on Image Information Processing (ICIIP) · doi:10.1109/iciip.2013.6707593 · Cited by 43
Quality Index for Stereoscopic Images by Jointly Evaluating Cyclopean Amplitude and Cyclopean Phase
Lin, Y., Yang, J., Lu, W. B., Meng, Q., Lv, Z., & Song, H. (2016) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2016.2632422 · Cited by 39
“In order to demonstrate its efficiency, the proposed method is compared with several existing state-of-art IQA metrics, including three 2D metrics (MS-SSIM [10], VIF [11] and ADM [60]), and four 3D metrics (Benoit’s scheme [61], Lin’s scheme [24], Chen’s scheme [28], Shao’s scheme-A [22] and Shao’s scheme-B [29]).”
BackgroundMethodInfluentialSCID: A database for screen content images quality assessment
Ni, Z.-K., Ma, L., Zeng, H., Fu, Y., Xing, L., & Ma, K.-K. (2017) · International Symposium on Intelligent Signal Processing and Communication Systems · doi:10.1109/ispacs.2017.8266580 · Cited by 38
“[2] [9] [10] [11] [12] [13] [14] [14] [15] [16] [17] [18] [18]”
MethodYue, G., Hou, C., Jiang, Q., & Yang, Y. (2018) · Signal Processing · doi:10.1016/j.sigpro.2018.04.019 · Cited by 39
General Resolution Enhancement Method in Atomic Force Microscopy Using Deep Learning
Liu, Y., Sun, Q., Lu, W., Wang, H., Sun, Y., Wang, Z., Lu, X., & Zeng, K. (2018) · Advanced Theory and Simulations · doi:10.1002/adts.201800137 · Cited by 44
Self-citationBurdziakowski, P. (2020) · Remote Sensing · doi:10.3390/rs12050810 · Cited by 43
Єремеєв, О., Лукин, В. В., Okarma, K., & Egiazarian, K. (2020) · Remote Sensing · doi:10.3390/rs12152349 · Cited by 46
“[89] 0.1094 very fast no MS-SSIM [66] 0.1139 very fast yes (only for 24 inputs) ESSIM [91] 0.1160 very fast no UQI [42] 0.1180 very fast no VIFP [74] 0.1277 very fast no HaarPSI [102] 0.1280 very fast yes PSIM [70] 0.1282 very fast no SFF [92] 0.1400 very fast no CSSIM [105] 0.1491 very fast no…”
BackgroundMethodInfluentialHyperrealistic Image Inpainting with Hypergraphs
Wadhwa, G., Dhall, A., Murala, S., & Tariq, U. (2021) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00396 · Cited by 54
“Nevertheless, we report, in Table-1, our evaluation results in terms of l 1 error, l 2 error, PSNR, MS-SSIM [48], and Frechet Inception Distance (FID) [18] on the validation set of places2 and testing set CelebA-HQ datasets.”
Pu, W. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3060747 · Cited by 32
“In the simulation, we have the ground truth of the separation results, and therefore, we can use PSNR and SSIM [69] as the evaluation metrics, which are Fig.”
MethodLee, K., Kim, J.-Y., Lee, M. H., Choi, C.-H., & Hwang, J. Y. (2021) · Sensors · doi:10.3390/s21062214 · Cited by 46
“Therefore, an advanced deep-learning algorithm further needs to be developed for the better diagnosis of RCT in ultrasound images [17–19].”
MethodClinical suitability of deep learning based synthetic CTs for adaptive proton therapy of lung cancer
Thummerer, A., Oria, C. S., Zaffino, P., et al. (2021) · Medical Physics · doi:10.1002/mp.15333 · Cited by 40
Luo, Q., Su, J., Yang, C., Gui, W., Sílven, O., & Liu, L. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3165270 · Cited by 55
“In addition, the hybrid loss l(k) is formulated as: ( ) ( ) ( ) ( )k k k k bce iou ssim l l l l= + + (8) Where lbce,liou and lssim represent binary cross entropy (BCE) [29], intersection-over-union (IoU) [30], structural similarity (SSIM) [31], respectively.”
BackgroundInfluentialLSVC: A Learning-based Stereo Video Compression Framework
Chen, Z., Lu, G. S., Hu, Z., Liu, S., Jiang, W., & Xu, D. (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/cvpr52688.2022.00598 · Cited by 46
“In addition, we use PSNR and MS-SSIM to measure the distortion between the compressed and original videos.”
MethodInfluentialSalt and Pepper Noise Detection and Removal in Gray Scale Images: An Experimental Analysis
Leavline, E. J., & Singh, D. (2013) · International Journal of Signal Processing Image Processing and Pattern Recognition · doi:10.14257/ijsip.2013.6.5.30 · Cited by 40
“Multi-scale structural similarity method and introduced in [16, 17] is an image synthesisbased approach to calibrate the parameters that weight the relative importance between different scales.”
MethodLee, J., Cho, S.-H., & Kim, M. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.12817 · Cited by 40
“2, and we additionally provide experimental results of the MS-SSIM [26] optimized models.”
MethodOszust, M. (2016) · PLOS ONE · doi:10.1371/journal.pone.0158333 · Cited by 39
Petascale pipeline for precise alignment of images from serial section electron microscopy
Popovych, S., Macrina, T., Kemnitz, N., Castro, M., Nehoran, B., Jia, Z., Bae, J. A., Mitchell, E., Mu, S., Trautman, E. T., et al. (2024) · Nature Communications · doi:10.1038/s41467-023-44354-0 · Cited by 39
Long-Term Temporally Consistent Unpaired Video Translation from Simulated Surgical 3D Data
Rivoir, D., Pfeiffer, M., Docea, R., Kolbinger, F. R., Riediger, C., Weitz, J., & Speidel, S. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00333 · Cited by 54
“Finally, we enforce a Multi-Scale Structural-Similarity loss [48, 37] Lssim on the brightness of a ref i and b̂i as well as b and â ref”
MethodFourier Document Restoration for Robust Document Dewarping and Recognition
Xue, C., Tian, Z., Zhan, F., et al. (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/cvpr52688.2022.00453 · Cited by 39
“We adopt two types of widely used evaluation metrics [8, 11, 27, 45] including: 1) Multi-Scale Structural Similarity (MS-SSIM) [43] and Local Distortion (LD) [45] that focus on image similarity performance; 2) Character Error Rate (CER) for evaluation of optical character recognition (OCR)…”
MethodNo-Reference Bitstream-Layer Model for Perceptual Quality Assessment of V-PCC Encoded Point Clouds
Liu, Q., Su, H., Chen, T., Yuan, H., & Hamzaoui, R. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3177926 · Cited by 42
“In addition to the aforementioned point-based models, there are also many projection-based models [38], [39], [42], [45], [46], [50], [53], such as projection-based PSNR (PSNRp) [38], [41], projection-based structural similarity (SSIMp) [9], [38], [41], projection-based multiscale structural similarity (MSSSIMp) [38],…”
Izumi, T., Amagasaki, M., Ishida, K., & Kiyama, M. (2022) · Journal of Water and Climate Change · doi:10.2166/wcc.2022.291 · Cited by 45
LFACon: Introducing Anglewise Attention to No-Reference Quality Assessment in Light Field Space
Qu, Q., Chen, X., Chung, Y. Y., & Cai, W. (2023) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2023.3247069 · Cited by 40
“On the one hand, numerous traditional metrics for 2D images have been proposed in the literature, including ERGAS [33], UQI [35], SSIM [36], multiscale SSIM [37], RASE [7], VSI [45], VIF [26], PSNR-B [43], and BRISQUE [22].”
Toward Extreme Image Compression With Latent Feature Guidance and Diffusion Prior
Li, Z., Zhou, Y., Hao, W., Ge, C., & Jiang, J. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3455576 · Cited by 115
“Meanwhile, we employ the Peak Signal-to-Noise Ratio ( PSNR ) and Multi-Scale Structural Similarity Index ( MS-SSIM ) [58] to measure the fidelity of reconstruction results.”
Lossy compression of images without visible distortions and its application
Lukin, V., Zriakhov, M., Ponomarenko, N., Krivenko, S., & Miao, Z. (2010) · IEEE 10th INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING PROCEEDINGS · doi:10.1109/icosp.2010.5655751 · Cited by 46
A New Image Quality Metric for Image Auto-denoising
Kong, X., Li, K., Yang, Q., Wenyin, L., & Yang, M. (2013) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2013.359 · Cited by 42
“Two maps N and P measuring the local structure similarity between the noisy image I and Mh and I and Îh are then computed based on SSIM, and the linear correlation coefficient of the two maps is used as an IQA metric.”
MethodInfluentialZhang, Y., & Chandler, D. M. (2013) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2001342 · Cited by 45
Comparison-Based Image Quality Assessment for Selecting Image Restoration Parameters
Liang, H., & Weller, D. S. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2601783 · Cited by 43
Perceptual Evaluation of Adversarial Attacks for CNN-based Image Classification
Fezza, S. A., Bakhti, Y., Hamidouche, W., & Olivier Déforges, A. (2019) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex.2019.8743213 · Cited by 53
Dual Generator Generative Adversarial Networks for Multi-domain Image-to-Image Translation
Tang, H., Xu, D., Wang, W., Yan, Y., & Sebe, N. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-20887-5_1 · Cited by 44
“To make the model benefit from multi-scale deep information, we refer to a multi-scale implementation of SSIM [45] which constrains SSIM over scales.”
A Filter for SAR Image Despeckling Using Pre-Trained Convolutional Neural Network Model
Pan, T., Dong, P., Yang, W. T., & Li, H.-C. (2019) · Remote Sensing · doi:10.3390/rs11202379 · Cited by 38
“Figure 8 gives the change trends of PSNR and SSIM with different L.”
BackgroundInfluentialHDR Reconstruction Based on the Polarization Camera
Wu, X., Zhang, H., Hu, X., Shakeri, M., Fan, C., & Ting, J. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.3005379 · Cited by 39
“With the ground truth HDR images being the reference output, we use PSNR and SSIM [17] as performance metrics to evaluate the competing methods.”
An Imaging Information Estimation Network for Underwater Image Color Restoration
Lu, J., Yuan, F., Yang, W., & Cheng, E. (2021) · IEEE Journal of Oceanic Engineering · doi:10.1109/joe.2021.3077692 · Cited by 41
Dual-Channel Multi-Task CNN for No-Reference Screen Content Image Quality Assessment
Zhang, C., Huang, Z., Liu, S., & Xiao, J. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3143321 · Cited by 44
Tian, Y., Pang, G., Liu, Y., Wang, C., Chen, Y., Liu, F., Singh, R., Verjans, J. W. H., Wang, M., & Carneiro, G. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-45676-3_2 · Cited by 61
Safety and Trustworthiness of Deep Neural Networks: A SurveyPreprint
Huang, X., Kroening, D., Kwiatkowska, M., Ruan, W., Sun, Y., Thamo, E., Wu, M., & Yi, X. (2018) · arXiv (Cornell University) · Cited by 40
Adaptive tri-plateau limit tri-histogram equalization algorithm for digital image enhancement
Paul, A. (2021) · The Visual Computer · doi:10.1007/s00371-021-02330-z · Cited by 38
“In quantitative assessment, quantitative measures like AIC [46], FSIM [47], MS-SSIM [48], VSI [49], and GMSD [50] are considered.”
BackgroundMethodInfluentialFPX-NIC: An FPGA-Accelerated 4K Ultra-High-Definition Neural Video Coding System
Jia, C., Hang, X., Wang, S., Wu, Y., Ma, S., & Gao, W. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3164059 · Cited by 51
“The distortion metrics can be signal level mean-square-error (MSE), || ˜ x i − x i || 22 or other perceptual based metrics [36]. x denotes the input patches which are normalized.”
BackgroundMethodDomain Knowledge Driven Semantic Communication for Image Transmission Over Wireless Channels
Li, A., Liu, X., Wang, G., et al. (2022) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2022.3216994 · Cited by 38
“We use SSIM [18] and learned perceptual image patch similarity (LPIPS) as the performance metrics [19].”
MethodMirror complementary transformer network for RGB‐thermal salient object detection
Jiang, X., Hou, Y., Tian, H., & Zhu, L. (2023) · IET Computer Vision · doi:10.1049/cvi2.12221 · Cited by 53
ABANet : Attention boundary‐aware network for image segmentation
Rezvani, S., Fateh, M., & Khosravi, H. (2024) · Expert Systems · doi:10.1111/exsy.13625 · Cited by 49
“In order to prevent division by zero, we empirically fixed C 1 ¼ 0 : 012 and C 2 ¼ 0 : 032 in this paper (Wang et al., 2003).”
Liao, X.-R., Wei, X.-K., Zhou, M.-L., Li, Z.-G., & Kwong, S. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3409176 · Cited by 41
Data Readiness for AI: A 360-Degree Survey
Hiniduma, K., Byna, S., & Bez, J. L. (2025) · ACM Computing Surveys · doi:10.1145/3722214 · Cited by 93
“To evaluate data volatility and timeliness, the authors replace a percentage of old instances with new ones in the training data while assuming a fixed currency.”
Phaseformer: Phase-Based Attention Mechanism for Underwater Image Restoration and Beyond
Khan, M. R., Negi, A., Kulkarni, A., Phutke, S. S., Vipparthi, S. K., & Murala, S. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00931 · Cited by 83
Cost-Oriented Adaptive Multimedia Delivery
Molnar, A., & Muntean, C. H. (2013) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2013.2244786 · Cited by 41
Automatic prediction of perceptual quality of multimedia signals—a survey
Seshadrinathan, K., & Bovik, A. C. (2010) · Multimedia Tools and Applications · doi:10.1007/s11042-010-0625-9 · Cited by 46
“[89].”
Self-citationOverview and Empirical Analysis of ISP Parameter Tuning for Visual Perception in Autonomous Driving
Yahiaoui, L., Horgan, J., Deegan, B. M., Yogamani, S. K., Eising, C., & Denny, P. E. (2019) · Journal of Imaging · doi:10.3390/jimaging5100078 · Cited by 52
“These included Universal Quality Index [17] and Structured Similarity (SSIM) [18], MS-SSIM [19], IFC [20], VIF [21], VSNR [22], FSIM [23] and salience weighted quality metrics [5].”
MethodInfluentialDeep Image Compression Using Decoder Side Information
Ayzik, S., & Avidan, S. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58520-4_41 · Cited by 38
“Evaluation criteria: Following [15,19,24,28] we evaluated our results by an averaged rate-distortion curve using MS-SSIM [33], which is reported to correlate better with human perception of distortion than mean squared error (MSE) and variants such as PSNR especially in cases where distortion is large [22,14].”
MethodTowards Learning-based Inverse Subsurface Scattering
Che, C., Luan, F., Zhao, S., Bala, K., & Gkioulekas, I. (2020) · IEEE International Conference on Computational Photography (ICCP) · doi:10.1109/iccp48838.2020.9105209 · Cited by 46
“We compare rendered and input images using L 2 error and MS-SSIM [69] (a benchmark perceptually-motivated image similarity metric).”
MethodOptical Flow and Mode Selection for Learning-based Video Coding
Ladune, T., Philippe, P., Hamidouche, W., Zhang, L., & Déforges, O. (2020) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp48831.2020.9287049 · Cited by 56
Hierarchical Predictive Coding-Based JND Estimation for Image Compression
Wang, H., Yu, L., Liang, J., Yin, H., Li, T., & Wang, S. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3037525 · Cited by 41
“The performance of the JND model is measured in terms of PSNR, Multi-scale structural similarity (MS-SSIM) [46] and the mean opinion score (MOS).”
MethodQuality Assessment of Image Super-Resolution: Balancing Deterministic and Statistical Fidelity
Zhou, W., & Wang, Z. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3547899 · Cited by 77
“The simplest FRmethods include PSNR, mean square error (MSE) and mean absolute error (MAE), which have been shown repeatedly in the literature to under-perform the family of structural similarity based models, which include the SSIM index [33], the multi-scale SSIM (MS-SSIM) index [36], the feature similarity (FSIM) i…”
MethodInfluentialSelf-citationQiu, D., Cheng, Y., & Wang, X. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2023.3272155 · Cited by 41
A MS-SSIM Optimal JPEG 2000 Encoder
Richter, T., & Kim, K. J. (2009) · DCC · doi:10.1109/dcc.2009.15 · Cited by 44
“…ˆ γ 2 j,b M j /2N j M b , finally, is equivalent to a fixed contrast sensitivity masking term, and contains both the relative amplitude gain due to the wavelet filtering process, as well as the band visibility term obtained by Bovik et al obtained by MS-SSIM calibration [16] with subjective tests.”
MethodAn image-computable psychophysical spatial vision model
Schütt, H. H., & Wichmann, F. A. (2017) · Journal of Vision · doi:10.1167/17.12.12 · Cited by 58
A multi-scale contrast-based image quality assessment model for multi-exposure image fusion
Lu, X., Cai, L., Zeng, H., Chen, J., Zhu, J., & Hou, J. (2017) · Signal Processing · doi:10.1016/j.sigpro.2017.12.013 · Cited by 38
Fixing a Broken ELBOPreprint
Alemi, A. A., Poole, B., Fischer, I., Dillon, J. V., Saurous, R. A., & Murphy, K. P. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.00464 · Cited by 496
“Generative compression models also have to work with both quantized latent spaces and approximately fixed decoder model families trained with perceptual losses such as MS-SSIM (Wang et al., 2003), which constrain the form of the learned distribution.”
MethodNTIRE 2021 Challenge on Quality Enhancement of Compressed Video: Methods and Results
Yang, R., Timofte, R., Liu, J., Xu, Y., Zhang, X., Zhao, M., Zhou, S., Chan, K. C. K., Zhou, S., Xu, X., et al. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00075 · Cited by 41
“They improve the average PSNR and MS-SSIM by 1.98 dB and 0.0257, respectively.”
MethodInfluentialKaza, N., Ojaghi, A., & Robles, F. E. (2022) · BME Frontiers · doi:10.34133/2022/9853606 · Cited by 39
“The MS-SSIM and SSIM were calculated for all three color channels of each image patch, and the values were averaged to obtain a single MS-SSIM or SSIM value for every patch.”
MethodInfluentialChai, X., Shao, F., Mu, B., Chen, H., Jiang, Q., & Ho, Y. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3350180 · Cited by 48
“…evaluation metrics for FR, RR, and NR are implemented as competitors followed by: 1) The fourteen FR methods including SSIM P [63], MS − SSIM P [64] and VIFP P [65] designed for 2D images, as well as PSNR MSE , po2po [7], PSNR HF , po2po [7], PSNR MSE , po2pl [11], PSNR HF , po2pl [11], PSNR Y…”
MethodSFCFusion: Spatial–Frequency Collaborative Infrared and Visible Image Fusion
Chen, H., Deng, L., Chen, Z., Liu, C., Zhu, L., Dong, M., Lu, X., & Guo, C. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3370752 · Cited by 58
“…evaluate different fusion methods, i.e., SSIM index [55], spatial frequency (SF) [56], transfer edge information quantity ( Q AB / F ) [57], pixel feature mutual information (FMI_pixel) [58], multiscale structural similarity measurement (MS-SSIM) [59], and difference correlation sum (SCD) [60].”
MethodA no-reference objective image quality metric based on perceptually weighted local noise
Zhu, T., & Karam, L. J. (2014) · EURASIP Journal on Image and Video Processing · doi:10.1186/1687-5281-2014-5 · Cited by 41
“In addition to the proposed FR-PWN and NRPWN metrics, the performance results of various existing metrics are presented for comparison, including seven full-reference metrics, DCTune [25], picture quality scale (PQS) [26], NQM [8], Fuzzy S7 [27], blockwise spectral distance measure (BSDM) [28], MS-SSIM [7],”
BackgroundMethodInfluentialFull-reference visual quality assessment for synthetic images: A subjective study
Kundu, D., & Evans, B. (2015) · International Conference on Information Photonics · doi:10.1109/icip.2015.7351227 · Cited by 42
“A (PHMA) [21] 0.661 0 0.713 1 0.859 0 0.842 2 0.896 1 0.806 2.8 12 Information Fidelity Criterion (IFC) [22] 0.728 2 0.792 0 0.837 0 0.913 0 0.850 3 0.791 8.0 13 Noise Quality Measure (NQM) [23] 0.753 0 0.837 0 0.880 0 0.919 0 0.859 2 0.790 2.4 14 Weighted Signal-to-Noise ratio (WSNR) [24] 0.617 1 0.745 0 0.845 1 0.87…”
Influential
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.