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 3,801–3,900 of 6,047
Blind image deblurring via L1-regularized second-order gradient prior
Liu, J., Tan, J., Zhang, L., Zhu, X., & Ge, X. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-13010-y · Cited by 3
“The performance evaluation standards such as error ratio [26], PSNR [47], SSIM [46], and kernel similarity are adopted to evaluate the proposed algorithm and other advanced methods.”
MethodInfluentialRosney, S., Donegan, C., Gower, M., et al. (2022) · doi:10.1117/12.2633075 · Cited by 3
Gu, F., & Zhang, Z. (2022) · Sensors · doi:10.3390/s22218084 · Cited by 3
“PSNR, SSIM [27], MS-SSIM [28] and BRISQUE [29] are 2D models.”
DIFAI: Diverse Facial Inpainting using StyleGAN Inversion
Yoon, D., Kwak, J.-G., Li, Y., et al. (2022) · Proceedings - International Conference on Image Processing · doi:10.1109/icip46576.2022.9898012 · Cited by 3
A Nested U-Shaped Residual Codec Network for Strip Steel Defect Detection
Guo, H., Zhan, S., Zhang, L., Zhu, W., Sun, Y., & Wang, J. (2022) · Applied Sciences · doi:10.3390/app122311967 · Cited by 3
Spiking Sparse Recovery With Non-Convex Penalties
Zhang, X., Yu, L., Zheng, G., & Eldar, Y. C. (2022) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2023.3234460 · Cited by 3
“We then employ the commonly used metrics peak signal-tonoise ratio (PSNR) and structural similarity (SSIM) [43] for quantitative evaluation.”
MethodInfluentialVideo Quality Assessment based on Quality Aggregation Networks
Wo, W., Zhang, Y., Hu, Y., Chen, Z., & Liu, S. (2022) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip56404.2022.10008817 · Cited by 3
Frequency-aware Learned Image Compression for Quality Scalability
Choi, H., Racapé, F., Hamidi-Rad, S., et al. (2022) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip56404.2022.10008818 · Cited by 3
“The last row shows coding gain by the models optimized for MS-SSIM in terms of BD-rate computed on the MS-SSIM versus bpp curves.”
A Full-reference Video Quality Assessment Method for 4K UHD Video based on Multi-Feature Fusion
Yi, G., Ping, S., & Pan, D. (2022) · ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD) · doi:10.1109/snpd54884.2022.10051773 · Cited by 4
“We compared the MFF-VQA model with the 6 existing FR objective quality assessment models, namely PSNR, FSIM, VIF, SSIM, MS-SSIM [30] and IFC [31].”
MethodKopriva, I., Shi, F., Lai, M., et al. (2023) · Physics in Medicine and Biology · doi:10.1088/1361-6560/acd6d1 · Cited by 3
Bi-attention network for bi-directional salient object detection
Xu, C., Wang, H., Liu, X., & Zhao, W. (2023) · Applied Intelligence · doi:10.1007/s10489-023-04648-8 · Cited by 4
HVS-inspired adversarial image generation with high perceptual quality
Xue, Y., Jin, J., Sun, W. B., & Lin, W. (2023) · Journal of Cloud Computing Advances Systems and Applications · doi:10.1186/s13677-023-00470-2 · Cited by 3
“…performance in various of traditional computer vision tasks, e.g., image classification [1], image processing [2–4],image quality assessment [5–8], etc. Research has demonstrated that deep learning algorithms can surpass human performance in specific tasks, such as facial recognition [9]…”
BackgroundQuality Assessment of Low-Light Restored Images: A Subjective Study and an Unsupervised Model
Kannan, V., Malik, S., Babu, N. C., & Soundararajan, R. (2023) · IEEE Access · doi:10.1109/access.2023.3292114 · Cited by 5
“the multi-scale processing of human vision [63].”
Image super-resolution techniques using deep neural networks
Guo, M. (2023) · Applied and Computational Engineering · doi:10.54254/2755-2721/5/20230567 · Cited by 4
Deep Learning-Based Machine Color Emotion Generation
Nie, T., & Lv, X.-G. (2023) · International Journal of Mobile Computing and Multimedia Communications · doi:10.4018/ijmcmc.325349 · Cited by 3
NRVC: Neural Representation for Video Compression with Implicit Multiscale Fusion Network
Liu, S., Cao, P., Feng, Y., et al. (2023) · Entropy · doi:10.3390/e25081167 · Cited by 3
“We evaluate the video quality with two metrics: PSNR and MS-SSIM [43].”
MethodWan, D., Jiang, X., & Shen, Q. (2023) · IEEE Access · doi:10.1109/access.2023.3312126 · Cited by 4
“Similar to 2D image quality assessment [3], [4], [5], [6],”
ResultFeature Structure Similarity Index for Hybrid Human and Machine Vision
Lin, Y., Wan, L., Ma, S., & Zhang, P. (2023) · Proceedings - International Conference on Image Processing · doi:10.1109/icip49359.2023.10222499 · Cited by 3
Luminance-Preserving Visible and Near-Infrared Image Fusion Network with Edge Guidance
Zhu, R., Ling, Y., Xiong, X., et al. (2023) · Proceedings - International Conference on Image Processing · doi:10.1109/icip49359.2023.10222294 · Cited by 3
“We adopt multiscale SSIM (MS-SSIM, or MS) [14] loss as our baseline.”
MethodDeep-Learning-Based Multitask Ultrasound Beamforming
Dahan, E., & Cohen, I. (2023) · Information · doi:10.3390/info14100582 · Cited by 3
“For both cases, our loss function was a linear combination of log l 1 and MS-SSIM [18] loss, given by: where ˆ y is the model envelope estimation, y is the ground-truth, and SSIM is the MS-SSIM loss.”
MethodBPG-based compression analysis of Poisson-noisy medical images
Науменко, В. В., Коваленко, Б., & Лукин, В. В. (2023) · RADIOELECTRONIC AND COMPUTER SYSTEMS · doi:10.32620/reks.2023.3.08 · Cited by 3
Venkata Phani Kumar, M., Ghosh, M., & Mahapatra, S. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-17228-2 · Cited by 3
Wang, Z., & He, G. (2024) · Sensors · doi:10.3390/s24010299 · Cited by 3
“We compute the multi-scale structural similarity index (MS-SSIM) [40] of adjacent frames for scene segmentation.”
MethodSelf-citationDashReStreamer: Framework for Creation of Impaired Video Clips under Realistic Network Conditions
Hodžić, K., Ćosović, M., Mrdović, S., Quinlan, J. J., & Raca, D. (2024) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3640016 · Cited by 5
“The most popular FR models DashReStreamer: Framework for Creation of Impaired Video Clips 15:3 are Peak Signal-to-Noise Ratio (PSNR) , Structural Similarity Index (SSIM) [84], Multiscale Structural Similarity Index (MS-SSIM) [85], and Video Multimethod Assessment Fusion (VMAF) [1].”
InfluentialA Near Lossless Learned Image Coding Network Quantization Approach for Cross-Platform Inference
Hang, X., Jia, C., Ma, S., & Gao, W. (2023) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip59821.2023.10402704 · Cited by 4
“1 https://github.com/LiuLei95/PyTorch-Learned-Image-Compression-with-GMM-and-Attention The bpp, PSNR, and MS-SSIM [21] results of (b) anchor method [1], two quantization methods, (c) AIMET [20] and (d) our approach are depicted.”
MethodPreprocessing-based Adversarial Defense for Object Detection via Feature Filtration
Zhou, L., Liu, Q., & Zhou, S. (2023) · International Conference on Algorithms, Computing and Systems (ICACS) · doi:10.1145/3631908.3631920 · Cited by 3
“In order to reconstruct the examples with high quality, a mixture of 𝑙 1 loss and MS-SIMM [39] loss is used to define the reconstruction loss: As for the decoder, we feed its outputs into the object detec-tor’s backbone 𝐵 .”
MethodLVC-LGMC: Joint Local and Global Motion Compensation for Learned Video Compression
Jiang, W., Li, J., Zhang, K., et al. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10448081 · Cited by 3
“The proposed models are optimized with the rate-distortion loss as follows, where D denotes the distortion metric Mean Square Error (MSE) or MS-SSIM [16].”
MethodCamPro: Camera-based Anti-Facial Recognition
Zhu, W., Sun, Y., Liu, J., Cheng, Y., Ji, X., & Xu, W. (2024) · Network and Distributed System Security Symposium · doi:10.14722/ndss.2024.24158 · Cited by 10
Self-Supervised Learning of Visual Robot Localization Using LED State Prediction as a Pretext Task
Nava, M., Carlotti, N., Crupi, L., et al. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3365973 · Cited by 3
“In AE-P, we train an autoencoder on T (cid:2) ∪ T u by minimizing the MS-SSIM [30] between input and reconstructed images; then, an additional NN head learns the localization task using L task on T (cid:2) , taking as input features computed by the autoencoder’s bottleneck.”
MethodReal-Time Monte Carlo Denoising With Adaptive Fusion Network
Lee, J., Lee, S., Yoon, M., et al. (2024) · IEEE Access · doi:10.1109/access.2024.3369588 · Cited by 3
Fu, H., Liang, F., Liang, J., et al. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10447420 · Cited by 3
Denoising OCT videos based on temporal redundancy
Richer, E., Solano, M. M., Farida Cheriet, A., Lesk, M. R., & Costantino, S. (2024) · Scientific Reports · doi:10.1038/s41598-024-56935-0 · Cited by 3
Learning Real-World Image De-weathering with Imperfect Supervision
Liu, X., Zhang, Z., Wu, X., Feng, C., Wang, X., Lei, L., & Zuo, W. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i4.28164 · Cited by 9
Automatic labeling of 3D facial acupoint landmarks
Yin, J., Fang, M., & Ma, W. (2024) · Metaverse · doi:10.54517/m.v5i1.2476 · Cited by 3
“We then calculate the MS-SSIM [42] similarity between the pattern and the average feature pattern and use the candidate point with the maximum similarity as the final optimized landmark.”
MethodAlkzir, N. B., Nikolaev, I. P., & Nikolaev, D. (2024) · doi:10.1117/12.3023509 · Cited by 3
“Multiscale structural similarity (MS-SSIM) [17], which is a modification of the SSIM metric that considers the images at several spatial scales at once: where 𝑗 is an index running through the considered scales.”
BackgroundRADIO: Reference-Agnostic Dubbing Video Synthesis
Lee, D., Kim, C., Yu, S., Yoo, J., & Park, G.-M. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00412 · Cited by 5
Analyzing the Quality of Distorted Images by the Normalized Mutual Information Measure
Haroutunian, M. E., Asatryan, D. G., & Mastoyan, K. A. (2024) · Mathematical Problems of Computer Science · doi:10.51408/1963-0111 · Cited by 3
“• (Higher Values, Better Quality) Higher PSNR values signify better image quality with minimal noise interference [2, 13, 14].”
BackgroundDigital Signatures for Authenticating Compressed JPEG Images
Erfurth, S. (2024) · doi:10.1145/3660512.3665522 · Cited by 6
“for compressed images.”
Gonzalez, J. A. C., Perez, K. D. J. R., Barros, T. T. L., Corso, G., & de Araújo, J. M. (2024) · Research Society and Development · doi:10.33448/rsd-v13i7.46226 · Cited by 3
Hsieh, A.-T., Chiu, C., Chen, T.-C., et al. (2024) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas58744.2024.10558157 · Cited by 3
“To address this, we train our FPN-ResUNet with Single-Scale Structural Similarity (SSIM) loss [32], which considers the perceptual similarity between image regions, aligning more closely with human visual perception.”
MethodFeature-level contrastive learning for full-reference light field image quality assessment
Lin, L., Qu, M., Bai, S., Wang, L., Wei, X., & Zhou, W. (2024) · Journal of the Franklin Institute · doi:10.1016/j.jfranklin.2024.107058 · Cited by 5
Yangue, E., Ranjan, A., Feng, Y., & Liu, C. (2024) · Journal of Medical Devices · doi:10.1115/1.4065905 · Cited by 3
“When SSIM is extended to measure structural similarity index using different image scales, termed multiscale SSIM MS-SSIM ð Þ [38], as shown in the following equation: where s is the number of scales.”
MethodFace Anti-Spoofing via Interaction Learning with Face Image Quality Alignment
Liu, Y., Li, Z., Li, S., Wang, Z., & Wu, L. (2024) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg59268.2024.10582010 · Cited by 3
“The first one directly incorporates multiple general image quality metrics (e.g., PSNR [20], SSIM [41] and UNIQUE [32]), as quality features to mine quality-based cues for FAS [5], [13], [15], [16].”
BackgroundSelf-citationElmeligy, B., Richter, T., Rao, R. R. R., et al. (2024) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex61742.2024.10598248 · Cited by 3
“These quality metrics are Peak Signal-to-Noise Ratio (PSNR), Multi-Scale Structural Similarity Index (MS-SSIM) [19], Deep Image Structure and Texture Similarity (DISTS) [20], Gradient Magnitude Similarity Deviation (GMSD) [21], Haar wavelet-based Perceptual Similarity Index (HaarPSI) [22], Visual…”
MethodYabuki, R., Yamaguchi, Y., Al-Ola, O. M. A., et al. (2024) · Journal of Imaging · doi:10.3390/jimaging10080178 · Cited by 3
“[29] and the PSNR between the true and reconstructed images.”
BackgroundZhang, S., Mao, W., Shi, H., et al. (2024) · doi:10.23919/date58400.2024.10546891 · Cited by 3
Self-citationLipowicz, P., Borowska, M., & Dardzińska, A. (2024) · Applied Sciences · doi:10.3390/app14167292 · Cited by 3
Karalasingham, S., Deo, R. C., Casillas-Pérez, D., Raj, N., & Salcedo‐Sanz, S. (2024) · Remote Sensing Applications Society and Environment · doi:10.1016/j.rsase.2024.101333 · Cited by 9
StyleCineGAN: Landscape Cinemagraph Generation Using a Pre-trained StyleGAN
Choi, J., Seo, K., Ashtari, A., & Noh, J. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.00752 · Cited by 12
“Quantitative Comparisons We compared the content preservation ability of both methods by measuring LPIPS [45], RMSE, and MS-SSIM [40] between the first frame ˆ I 0 and the subsequent frames ˆ I n .”
MAGICK: A Large-Scale Captioned Dataset from Matting Generated Images Using Chroma Keying
Burgert, R., Price, B. L., Kuen, J., Li, Y., & Ryoo, M. S. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.02132 · Cited by 20
“We measure this similarity using MSS-SIM [39] (multi-scale structural image similarity) as we found it gave the best results empirically.”
MethodDeep learning-based prediction of later 13N-ammonia myocardial PET image frames from initial frames
Mokri, M., Safari, M., Kaviani, S. I., Juneau, D., Cohalan, C., Archambault, L., & Carrier, J. (2024) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2024.106865 · Cited by 3
Real-Time Video Prediction With Fast Video Interpolation Model and Prediction Training
Hirose, S., Kotoyori, K., Arunruangsirilert, K., Lin, F., Sun, H., & Katto, J. (2024) · Proceedings - International Conference on Image Processing · doi:10.1109/icip51287.2024.10647865 · Cited by 4
“[23] is used as the evaluation metrics.”
Hongde, 张. Z., Xin, 冯. F., Jieming, 杨. Y., et al. (2024) · ACTA PHOTONICA SINICA · doi:10.3788/gzxb20245308.0810004 · Cited by 3
MLP Embedded Inverse Tone Mapping
Liu, P., Li, J., Wang, L., Zha, Z.-J., & Xiong, Z. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3680937 · Cited by 3
DeepEraser: Deep Iterative Context Mining for Generic Text Eraser
Feng, H., Wang, W., Liu, S., Deng, J., Zhou, W., & Li, H. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3521809 · Cited by 9
“Signal-to-Noise Ratio (PSNR), Multi-scale Structure Similarity Index Measure (MSSIM) [50], Mean Squared Error (MSE), along with AGE, pEPs, and pCEPS [13].”
Fu, Y., Zhao, Z., Ji, Y., & Quek, T. Q. S. (2024) · IEEE Communications Letters · doi:10.1109/lcomm.2024.3523545 · Cited by 3
“• Multi-scale structural similarity (MS-SSIM) in [14]: It is a structural perceptual metric, which measures the similarity of two images at multiple scales by calculating luminance, contrast, and structure.”
Ng, Y. J., & Sim, K. S. (2024) · International Journal on Advanced Science Engineering and Information Technology · doi:10.18517/ijaseit.14.6.10115 · Cited by 3
“For the multi-scale histogram equalization [58], we can use histogram equalization across a range of resolutions or scales.”
Optimizing Natural Image Quality Evaluators for Quality Measurement in CT Scan Denoising
Gunawan, R., Tran, Y., Zheng, J., et al. (2025) · Computers · doi:10.3390/computers14010018 · Cited by 3
Combining Pre- and Post-Demosaicking Noise Removal for RAW Video
Sánchez-Beeckman, M., Buades, A., Brandonisio, N., & Kanoun, B. (2025) · IEEE Transactions on Image Processing · doi:10.1109/tip.2025.3527886 · Cited by 4
“We extract the PSNR, SSIM [104], MS-SSIM [105], and VMAF [103] scores from the output of the software.”
Enhancing Learned Image Compression via Cross Window-Based Attention
Mudgal, P., & Liu, F. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-77389-1_32 · Cited by 5
Tang, W.-Y., Xiang, Z., Yu, S., Zhi, J., & Yang, Z. (2025) · Journal of Engineering Design · doi:10.1080/09544828.2025.2450760 · Cited by 3
“MS-SSIM further considers structural similarity at different based on structural similarity index measure (SSIM), with values ranging from 0 to 1.”
InfluentialMVCD: Multi-Dimensional Video Compression Dataset
Amirpour, H., Ghasempour, M., Tashtarian, F., et al. (2024) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip63160.2024.10849883 · Cited by 3
“For each video sequence, the spatial and temporal complexity, bitrate, and video quality metrics ( i.e. , Peak signal-to-noise ratio (PSNR) [21], Structural Similarity (SSIM), Multi-scale SSIM [22], and VMAF) are provided.”
A Novel Lightweight Image Cryptographic Algorithm via Substitution‐Permutation Methods
Gupta, T., Selwal, A., & Sharma, A. K. (2025) · Security and Privacy · doi:10.1002/spy2.70015 · Cited by 5
Transfer Learning and Double U-Net Empowered Wave Propagation Model in Complex Indoor Environments
Fu, Z., Mukherjee, S., Lanagan, M. T., et al. (2025) · IEEE Transactions on Antennas and Propagation · doi:10.1109/tap.2025.3553952 · Cited by 3
“The Pearson correlation is calculated as In addition, the structural similarity index (SSIM) [35], a measurement of the visual similarity between two images by considering changes in luminance, contrast, and structure, is applied to evaluate the radio map predictions by the trained model.”
Enhancing Few-Shot SAR Ship Recognition: Pseudospectrum Information Generation and Fusion
Gao, G., Liu, W., & Zhang, X. (2025) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2025.3561516 · Cited by 5
Zhang, Z., Meng, C.-B., Jiang, X.-L., Zhao, C., Qiao, S., & Zhang, T. (2025) · Nuclear Science and Techniques · doi:10.1007/s41365-025-01686-4 · Cited by 7
“For the feature fusion model, the mean and standard deviation of the feature vectors were selected as the most representative features [29, 34, 35].”
Lili, 张. Z., Xiangru, 栗. L., Jiahui, 刘. L., & Qizhi, 房. F. (2025) · Chinese Journal of Lasers · doi:10.3788/cjl250499 · Cited by 3
Multilevel retentive networks with edge enhanced attention for infrastructure crack detection
Zou, Z., Yang, S., Wang, M., & Song, B. (2025) · Engineering Structures · doi:10.1016/j.engstruct.2025.121191 · Cited by 3
Diffusion Bridge Models for 3D Medical Image Translation
Zhang, S., Chattopadhyay, T., Thomopoulos, S. I., Ambite, J. L., Thompson, P. M., & Steeg, G. V. (2025) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc58623.2025.11252839 · Cited by 4
“We used the Multi-Scale Structural Similarity Index Measure (MS-SSIM) [31], as implemented in the MONAI generative module, to evaluate perceptual similarity between generated and reference images.”
Optimizing the image R/D coding performance by tuning quantization parameters
Martínez-Rach, M., Piñol, P., López-Granado, O., & Malumbres, M. P. (2017) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2017.09.015 · Cited by 3
No reference image quality assessment: Feature fusion using relevance vector machine
Sadou, B., Lahoulou, A., & Bouden, T. (2017) · International Conference on Electrical Engineering - Boumerdes (ICEE-B) · doi:10.1109/icee-b.2017.8192196 · Cited by 4
“Some other FR algorithms have been established and proven to correlate well with human perception of quality such as structural similarity index (SSIM) [2], multi scale structural similarity index (MS-SSIM) [3] the visual information fidelity (VIF) [4], the visual signalto-noise ratio (VSNR) [5].”
BackgroundNo-Reference Contrast Measurement for Color Images Based on Visual Stimulus
Kim, M., Song, K. S., & Kang, M. G. (2018) · IEEE Access · doi:10.1109/access.2018.2828830 · Cited by 4
“The popular algorithms are peak signal to noise ratio (PSNR) [2], structural similarity index measure (SSIM) [3], [4] and using the machine learning [5], or the neural network [6].”
MethodLi, C., Li, Y., Yuan, Y., et al. (2018) · PLoS ONE · doi:10.1371/journal.pone.0199430 · Cited by 3
XmoNet: A Fully Convolutional Network for Cross-Modality MR Image Inference
Bano, S., Asad, M., Fetit, A. E., & Rekik, I. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-00320-3_16 · Cited by 3
“PSNR and SIMM were”
BackgroundMethodInfluentialNian, Z., & Jung, C. (2018) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2018.8451302 · Cited by 3
“That is, the MSE ability of capturing perceptually relevant differences was very limited because it was only based on pixel-wise image differences [19, 20, 21, 22].”
BackgroundImproving deep image denoising using pseudo‐ground‐truth images
Jung, S. (2019) · Electronics Letters · doi:10.1049/el.2019.1605 · Cited by 3
“Therefore, many alternative loss functions such as L1-loss, structural similarity index (SSIM), and multi-scale SSIM (MS-SSIM) [2] have been investigated for image denoising [3].”
BackgroundNo-Reference Image Quality Assessment Based on Multi-scale Convolutional Neural Networks
Chen, P., Niu, Y., & Huang, D. (2019) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-22868-2_84 · Cited by 2
Matching The Characteristics Of Fundus And Smartphone Camera Images
Vasudeva, S. A., & Sivaswamy, J. (2019) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi.2019.8759381 · Cited by 2
Quality Image Enhancement from Low Resolution Camera using Convolutional Neural Network
Patmawati, N. P., Arifianto, A., & Ramadhani, K. N. (2019) · doi:10.1109/icoict.2019.8835273 · Cited by 3
“While to calculate the SSIM value can be used [19], [20]:”
MethodA Neural Network Model of Spatial Distortion Sensitivity for Video Quality Estimation
Becker, S., Müller, K., Wiegand, T., & Bosse, S. (2019) · Machine learning for signal processing ... · doi:10.1109/mlsp.2019.8918899 · Cited by 3
“Classical approaches either imitate specific characteristics of the human visual system [1, 2, 3, 4] or measure distortions in the natural image statistics [5, 6, 7].”
BackgroundMethodDeep Feature Disentanglement Learning for Bone Suppression in Chest Radiographs
Lin, C., Tang, R., Lin, D. D., et al. (2020) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi45749.2020.9098399 · Cited by 3
“In order to demonstrate the effectiveness of the proposed architecture and the associated loss function, we trained four different bone suppression models: original autoencoder, autoencoder with MS-SSIM loss, disentanglement autoencoder, and disentanglement autoencoder with MS-SSIM loss.”
MethodInfluentialImage Noise Level Classification Technique Based on Image Quality Assessment
Luo, G., Zicheng, Z., Long, Q., Chun, L., & Bao, J. (2020) · IEEE International Conference on Power, Intelligent Computing and Systems (ICPICS) · doi:10.1109/icpics50287.2020.9202118 · Cited by 5
“Structural Similarity (SSIM) is used to measure the similarity between two images [7].”
MethodCuriously Effective Features For Image Quality Prediction
Becker, S., Wiegand, T., & Bosse, S. (2021) · Proceedings - International Conference on Image Processing · doi:10.1109/icip42928.2021.9506756 · Cited by 5
“One of the most prominent and seminal examples for scene statistic-based FR quality estimation is SSIM [3] that inspired many other models [4, 5].”
BackgroundRajagopal, H., Mokhtar, N. B., Khairuddin, A. S. M., et al. (2021) · Proceedings of International Conference on Artificial Life and Robotics · doi:10.5954/icarob.2021.os1-1 · Cited by 3
Stereoscopic Video Quality Assessment in the Context of Internet of Things
Zhao, Y. S., Yang, J., & Yang, S. (2021) · IEEE Consumer Electronics Magazine · doi:10.1109/mce.2021.3097305 · Cited by 3
Rani, P., Jain, V., Saif, M., et al. (2021) · Advances in intelligent systems and computing · doi:10.1007/978-981-16-3071-2_6 · Cited by 3
Color Image Quality Evaluation based on Visual Saliency and Gradient Information
Chang, H.-W., Du, C.-Y., Bi, X.-D., & Wang, M. (2021) · International Symposium on System and Software Reliability (ISSSR) · doi:10.1109/isssr53171.2021.00030 · Cited by 4
Real-time CPU-based View Synthesis for Omnidirectional Video
Stankowski, J., & Dziembowski, A. (2022) · Computer Science Research Notes · doi:10.24132/csrn.3201.30 · Cited by 3
Depth Patch Selection for Decoder-Side Depth Estimation in MPEG Immersive Video
Milovanović, M., Henry, F., & Cagnazzo, M. (2022) · Picture Coding Symposium · doi:10.1109/pcs56426.2022.10018042 · Cited by 3
“BD-rate is calculated in terms of Y-PSNR, VMAF [21], MS-SSIM [22], and IVPSNR [23], where IV-PSNR is a metric adapted for specific characteristics of immersive video.”
MethodInfluentialAnalyzing the Impact of Lossy Data Reduction on Volume Rendering of Cosmology Data
Wang, J., Grosset, P., Turton, T. L., & Ahrens, J. P. (2022) · IEEE/ACM International Workshop on Data Analysis and Reduction for Big Scientific Data (DRBSD) · doi:10.1109/drbsd56682.2022.00007 · Cited by 3
“…human perception into account by including effects such as luminance (feature similarity index, FSIM [19]), more heavily weighting edge effects (multi-component structural similarity index, 3DSSIM [20]), information content weighting (IW-SSIM [21]), or a multi-scale weighted SSIM (MS-SSIM [22]).”
BackgroundQuality assessment of adaptive bitrate videos using image metrics and machine learning
Søgaard, J., Forchhammer, S. O., & Brunnstrom, K. E. (2015) · International Conference on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex.2015.7148105 · Cited by 3
The latest challenges and opportunities in the current single image dehazing algorithms
Mai, J., Zhu, Q., & Wu, D. (2014) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio.2014.7090317 · Cited by 4
“UQI [31] and SSIM [32, 33] are used to measure the structural similarity between the input hazy image and the dehazing result.”
MethodA parametric video quality model based on source and network characteristics
Zerman, E., Konuk, B., Yılmaz, G. N., & Akar, G. B. (2014) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2014.7025119 · Cited by 3
Priya, K., Manasa, K., & Channappayya, S. S. (2014) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2014.6854108 · Cited by 3
Online learning for quality-driven unequal protection of scalable video
Khalek, A. A., Caramanis, C., & Heath, R. W. (2012) · IEEE International Workshop on Machine Learning for Signal Processing (MLSP) · doi:10.1109/mlsp.2012.6349781 · Cited by 3
“For quality assessment, we use the multiscale SSIM index [8] which was shown to provide very good correlation with subjective quality.”
MethodSCALE-SELECTIVE VALIDATION OF MORPHODYNAMIC MODELS
Bosboom, J., & Reniers, A. (2014) · Coastal Engineering Proceedings · doi:10.9753/icce.v34.sediment.75 · Cited by 3
“Also, in the field of image processing, Wang et al. (2004) determine the closeness of images using a multi-scale method, which incorporates image details at different resolutions.”
MethodEfficiency of lossy compression of noisy and pre-filtered remote sensing images
Лукин, В. В., Zemliachenko, A., & Tchobanou, M. K. (2013) · International Kharkiv Symposium on Physics and Engineering of Microwaves, Millimeter and Submillimeter Waves · doi:10.1109/msmw.2013.6622050 · Cited by 3
About the imperfection of objective quality metrics on high-definition video content
Wulf, S., & Zölzer, U. (2013) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp.2013.6659319 · Cited by 3
Perceptual non local mean (P-NLM) denoising
Souidene, W., Megrhi, S., Beghdadi, A., et al. (2012) · International Symposium on Communications, Control and Signal Processing · doi:10.1109/isccsp.2012.6217809 · Cited by 3
Evaluation of floating point image compression
Richter, T. (2009) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2009.5413692 · Cited by 3
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.