{"id":19250602,"url":"https://github.com/superbrucejia/nlnet-iqa","last_synced_at":"2025-06-27T11:40:27.660Z","repository":{"id":130813790,"uuid":"521105624","full_name":"SuperBruceJia/NLNet-IQA","owner":"SuperBruceJia","description":"Non-local Modeling for Image Quality Assessment","archived":false,"fork":false,"pushed_at":"2023-12-20T22:16:26.000Z","size":17083,"stargazers_count":13,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-21T12:44:12.281Z","etag":null,"topics":["blind","blind-image-quality-assessment","convolutional-neural-networks","graph-neural-networks","image-processing","image-quality","image-quality-assesment","image-quality-assessment","iqa","local-modeling","no-reference","no-reference-image-quality-assessment","non-local","non-local-modeling","non-local-nets","non-local-network","vggnet"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/SuperBruceJia.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2022-08-04T03:08:42.000Z","updated_at":"2024-11-06T14:16:04.000Z","dependencies_parsed_at":"2023-12-21T00:59:53.907Z","dependency_job_id":null,"html_url":"https://github.com/SuperBruceJia/NLNet-IQA","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/SuperBruceJia/NLNet-IQA","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SuperBruceJia%2FNLNet-IQA","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SuperBruceJia%2FNLNet-IQA/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SuperBruceJia%2FNLNet-IQA/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SuperBruceJia%2FNLNet-IQA/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/SuperBruceJia","download_url":"https://codeload.github.com/SuperBruceJia/NLNet-IQA/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SuperBruceJia%2FNLNet-IQA/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":262248915,"owners_count":23281862,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["blind","blind-image-quality-assessment","convolutional-neural-networks","graph-neural-networks","image-processing","image-quality","image-quality-assesment","image-quality-assessment","iqa","local-modeling","no-reference","no-reference-image-quality-assessment","non-local","non-local-modeling","non-local-nets","non-local-network","vggnet"],"created_at":"2024-11-09T18:18:01.574Z","updated_at":"2025-06-27T11:40:27.602Z","avatar_url":"https://github.com/SuperBruceJia.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Non-local Modeling for Image Quality Assessment\n\u003cimg width=\"1680\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/212229645-041c280b-fcaf-4d5a-8703-2d94f2fe1615.png\"\u003e\n\n## Table of Contents\n\u003cul\u003e\n    \u003cli\u003e\u003ca href=\"#Installation\"\u003eInstallation\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Experiments-Settings-and-Quick-Start\"\u003eExperiments Settings and Quick Start\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Superpixel-Segmentation-Demo\"\u003eSuperpixel Segmentation Demo\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Trained-Models-and-Benchmark-Databases\"\u003e[Download] Trained Models and Benchmark Databases\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Evaluation-Metrics\"\u003eEvaluation Metrics\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Motivation\"\u003eMotivation\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Local-Modeling-and-Non-local-Modeling\"\u003e[Definition] Local Modeling and Non-local Modeling\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Global-Distortions-and-Local-Distortions\"\u003e[Definition] Global Distortions and Local Distortions\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Paper-and-Presentations\"\u003e[Download] Paper and Presentations\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Structure-of-the-Code\"\u003eStructure of the Code\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Citation\"\u003eCitation\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Contact\"\u003eContact\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003ca href=\"#Acknowledgement\"\u003eAcknowledgement\u003c/a\u003e\u003c/li\u003e\n\u003c/ul\u003e\n\n## Installation\nFramework: PyTorch, OpenCV, PIL, scikit-image, scikit-learn, Numba JIT, Matplotlib, etc.\u003cbr\u003e\n**Note**: The overall framework is based on **PyTorch**. Here, I didn't provide a specific `pip install -r requirements.txt` because there are so many dependencies. I would like to suggest you install the corresponding packages when they are required to run the code.\n\n## Experiments Settings and Quick Start\n### Intra-Database Experiments\nExperiments Settings: 👉 Check [this file](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/lib/make_index.py#L8)\u003cbr\u003e\n✔︎ Split the reference images into 60% training, 20% validation, and 20% testing.\u003cbr\u003e\n✔︎ 10 random splits of the reference indices by setting seed `random.seed(random_seed)` from 1 to 10 `args.exp_id`.\u003cbr\u003e\n✔︎ The median SRCC and PLCC on the testing set are reported.\u003cbr\u003e\n\nQuick Start:\u003cbr\u003e\n```python\npython main.py --database_path '/home/jsy/BIQA/' --database TID2013 --batch_size 4 --num_workers 8 --gpu 0\n```\n(1) Other hyper-parameters can also be modified via `--parameter XXX`, _e.g._, `--epochs 200` and `--lr 1e-5`.\u003cbr\u003e\n(2) Hyper-parameters can be found from the `parser` in the [main.py](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/main.py#L73).\u003cbr\u003e\n(3) Please change the database path `'/home/jsy/BIQA/'` to your own path.\n\n\u003cdetails\u003e\n\u003csummary\u003eExperimental Results\u003c/summary\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211454477-1f112208-6f3f-45fe-8cfc-86fb311e243a.png\"\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211454790-904f12cb-ae83-4bb2-8eea-bd762b64c0f4.png\"\u003e\n\u003c/details\u003e\n\n### Cross-Database Evaluations\nExperiments Settings: 👉 Check [this file](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/Cross%20Database%20Evaluations/data_process/get_data.py#L50)\u003cbr\u003e\n✔︎ One database is used as the training set, and the other databases are the testing sets.\u003cbr\u003e\n✔︎ The performance of the model in the last epoch (100 epochs in this work) is reported.\u003cbr\u003e\n\nQuick Start: (Folder: Cross Database Evaluations)\u003cbr\u003e\n```python\npython cross_main.py --database_path '/home/jsy/BIQA/' --train_database TID2013 --test_database CSIQ --num_workers 8 --gpu 0\n```\n\n\u003cdetails\u003e\n\u003csummary\u003eExperimental Results\u003c/summary\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211455285-b29db97a-29d8-499a-a728-5707afe56e22.png\"\u003e\n\u003c/details\u003e\n\n### Single Distortion Type Evaluation\nQuick Start (Folder: Individual Distortion Evaluation):\n```python\npython TID2013-Single-Distortion.py\n```\n(1) Please change the trained models' path and Database path.\u003cbr\u003e\n(2) The Index of Distortion Type can be found from original papers: [TID2013](https://www.sciencedirect.com/science/article/pii/S0923596514001490) and [KADID](http://database.mmsp-kn.de/kadid-10k-database.html#:~:text=blurs). \n\n\u003cdetails\u003e\n\u003csummary\u003eExperimental Results\u003c/summary\u003e\n\nLIVE Database:\n\n\u003cimg width=\"700\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211454955-d9346292-b718-45f5-8f8a-14c81cc19586.png\"\u003e\n\n---\n\nCSIQ Database:\n\n\u003cimg width=\"700\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211455036-99a31158-967d-46b4-8ba1-4a2187447373.png\"\u003e\n\n---\n\nTID2013 Database:\n\n\u003cimg width=\"1400\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211455110-c48a94ca-599c-45a5-97e7-4d735cd994e5.png\"\u003e\n\n---\n\nKADID-10k Database:\n\n\u003cimg width=\"1400\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211455189-c367264c-03c5-49d0-8388-e8cdb1de6a49.png\"\u003e\n\u003c/details\u003e\n\n### Real World Image Testing\nQuick Start:\n```python\npython real_testing.py --model_file 'save_model/TID2013-32-4-1.pth' --im_path 'test_images/cr7.jpg' --database TID2013\n```\nPlease comment [these lines](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/real_testing.py#L45) if you don't want to resize the original image.\n\n## Superpixel Segmentation Demo\nQuick Start (Folder: Superpixel Segmentation):\n```python\npython superpixel.py\n```\n\n\u003cdetails\u003e\n\u003csummary\u003eSuperpixel vs. Square Patch Representation Demo\u003c/summary\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/210959208-6381e2f1-0b0f-4bd6-90b2-8a2039c08a09.png\"\u003e\n\u003cimg width=\"1600\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/212520023-5d774076-d10a-4cdc-b773-0115e4bd1c81.png\"\u003e\n\u003cimg width=\"1600\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211976748-3cbae528-d4e8-4f7a-893f-f04110e36abe.png\"\u003e\n\u003cimg width=\"1600\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211976777-e617f93f-9237-407d-bdcf-36874f4c66d4.png\"\u003e\n\u003c/details\u003e\n\n## Trained Models and Benchmark Databases\nAll trained models and benchmark databases are available on 🤗 [Hugging Face](https://huggingface.co/shuyuej/NLNet/tree/main).\\\n✔︎ Trained Models (Intra-Database Experiments): Download [here](https://drive.google.com/drive/folders/1K-24RGXyvSUZfnTThQ0CXUf4BgJA_pn7?usp=sharing)\u003cbr\u003e\n✔︎ Trained Models (Cross-Database Evaluations): Download [here](https://drive.google.com/drive/folders/1-9XfTt4ne057Ureecf_eLXiMQ_4xucgJ?usp=sharing)\u003cbr\u003e\n✔︎ LIVE, CSIQ, TID2013, and KADID-10k Databases: Download [here](https://drive.google.com/drive/folders/1gfBlByg1bpBXQOFZb6LyCttaX4eAf_Eh?usp=sharing)\n\n\u003cdetails\u003e\n\u003csummary\u003eDatabases Summary\u003c/summary\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211455700-1436735c-eec6-4670-b509-2bf784a11aee.png\"\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211731935-e1559cf9-15aa-4a33-bf86-540deb70028a.png\"\u003e\n\u003c/details\u003e\n\n## Evaluation Metrics\n(1) Pearson Linear Correlation Coefficient (**PLCC**): measures the prediction accuracy\u003cbr\u003e\n(2) Spearman Rank-order Correlation Coefficient (**SRCC**): measures the prediction monotonicity\u003cbr\u003e\n✔︎ A short note of the IQA evaluation metrics can be downloaded [here](https://shuyuej.com/files/MMSP/IQA_Evaluation_Metrics.pdf).\u003cbr\u003e\n✔︎ In the [code](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/lib/utils.py#L29) (`evaluation_criteria` function), PLCC, SRCC, Kendall Rank-order Correlation Coefficient (KRCC), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Outlier Ratio (OR) are all calculated. In this work, I only compare the PLCC and SRCC among different IQA algorithms.\n\n## Motivation\n**Local Content**: HVS is adaptive to the local content.\u003cbr\u003e\n**Long-range Dependency and Relational Modeling**: HVS perceives image quality with long-range dependency constructed among different regions.\n\n\u003cimg width=\"1460\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/212519878-0da16724-750d-43a2-a083-fc593463ad43.png\"\u003e\n\u003cimg width=\"1460\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/212519886-c7fb7ca1-ec50-4b7d-8e21-2287e00cf29c.png\"\u003e\n\n## Local Modeling and Non-local Modeling\n**Local Modeling**: The local modeling methods encode spatially proximate local neighborhoods.\u003cbr\u003e\n**Non-local Modeling**: The non-local modeling establishes the spatial integration of information by long- and short-range communications with different spatial weighting functions.\n\n\u003cimg width=\"1460\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/212520317-88da5cb5-44be-41d5-bcbe-1632c2c35811.png\"\u003e\n\u003cimg width=\"1460\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/212519914-d3948de4-dc2e-4125-9414-9725aa76af03.png\"\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003eNon-local Behavior Demo\u003c/summary\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211026397-7990fdbd-b41a-414a-a40f-ec4ecb637dcf.png\"\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211026979-60e49649-75c5-481f-86d0-021c2ad5cde6.png\"\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003eLocal Modeling vs. Non-local Modeling Demo\u003c/summary\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/211028273-373e8139-f111-40be-b214-019a64b90392.png\"\u003e\n\u003c/details\u003e\n\n## Global Distortions and Local Distortions\n**Global Distortions**: the globally and uniformly distributed distortions with non-local recurrences over the image.\u003cbr\u003e\n**Local Distortions**: the local nonuniform-distributed distortions in a local region.\n\n\u003cimg width=\"1460\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/212519962-20781700-29a6-4a2f-97a9-13beaaf8ac72.png\"\u003e\n\n\u003cimg width=\"1460\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/212447958-f8011613-e26b-4bf4-993a-b56c395703b6.png\"\u003e\n✔︎ LIVE Database:\n    \n    Global Distortions: JPEG, JP2K, WN, and GB\n    \n    Local Distortions: FF\n\n\u003cdetails\u003e\n\u003csummary\u003eDistortion Demo\u003c/summary\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/210927999-48d2d4e2-d63a-4ece-8681-e5fbe1fb3d98.png\"\u003e\n\u003c/details\u003e\n\n✔︎ CSIQ Database:\n    \n    Global Distortions: JPEG, JP2K, WN, GB, PN, and СС\n    \n    Local Distortions: There is no local distortion in CSIQ Database.\n\n\u003cdetails\u003e\n\u003csummary\u003eDistortion Demo\u003c/summary\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/210928260-3f3d938e-53eb-43fb-90d2-7cca59850aee.png\"\u003e\n\u003c/details\u003e\n\n✔︎ TID2013 Database:\n    \n    Global Distortions: Additive Gaussian noise, Lossy compression of noisy images, Additive noise in color components, Comfort noise, Contrast change, Change of color saturation, Spatially correlated noise, High frequency noise, Impulse noise, Quantization noise, Gaussian blur, Image denoising, JPEG compression, JPEG 2000 compression, Multiplicative Gaussian noise, Image color quantization with dither, Sparse sampling and reconstruction, Chromatic aberrations, Masked noise, and Mean shift (intensity shift)\n    \n    Local Distortions: JPEG transmission errors, JPEG 2000 transmission errors, Non eccentricity pattern noise, and Local bock-wise distortions with different intensity\n\n\u003cdetails\u003e\n\u003csummary\u003eDistortion Demo\u003c/summary\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/210928535-4f9bc8ad-f9ca-4a25-bbc0-5a9d3016a637.png\"\u003e\n\u003c/details\u003e\n\n✔︎ KADID-10k Database:\n    \n    Global Distortions: blurs (lens blur, motion blur, and GB), color distortions (color diffusion, color shift, color saturation 1, color saturation 2, and color quantization), compression (JPEG and JP2K), noise (impulse noise, denoise, WN, white noise in color component, and multiplicative noise), brightness change (brighten, darken, and mean shift), spatial distortions (jitter, pixelate, and quantization), and sharpness and contrast (high sharpen and contrast change)\n    \n    Local Distortions: Color block and Non-eccentricity patch\n\n\u003cdetails\u003e\n\u003csummary\u003eDistortion Demo\u003c/summary\u003e\n\u003cimg width=\"900\" alt=\"image\" src=\"https://user-images.githubusercontent.com/31528604/210928749-1d080cc4-04b4-462e-bc3b-0e6e3344d38d.png\"\u003e\n\u003c/details\u003e\n\n## Paper and Presentations\n(1) **Thesis** can be downloaded [here](https://scholars.cityu.edu.hk/en/theses/noreference-image-quality-assessment-via-nonlocal-modeling(2d1e72fb-2405-43df-aac9-4838b6da1875).html).\u003cbr\u003e\n(2) **Original Paper** can be downloaded [here](https://shuyuej.com/files/MMSP/MMSP22_Paper.pdf).\u003cbr\u003e\n(3) **Detailed Slides Presentation** can be downloaded [here](https://shuyuej.com/files/Presentation/A_Summary_Three_Projects.pdf).\u003cbr\u003e\n(4) **Detailed Slides Presentation with Animations** can be downloaded [here](https://shuyuej.com/files/Presentation/A_Summary_Three_Projects_Animations.pdf).\u003cbr\u003e\n(5) **Simple Slides Presentation** can be downloaded [here](https://shuyuej.com/files/MMSP/MMSP22_Slides.pdf).\u003cbr\u003e\n(6) **Poster Presentation** can be downloaded [here](https://shuyuej.com/files/MMSP/MMSP22_Poster.pdf).\n\n### Model Overiew\n\u003cdiv\u003e\n    \u003cdiv style=\"text-align:center\"\u003e\n    \u003cimg width=100%device-width src=\"https://github.com/SuperBruceJia/NLNet-IQA/raw/main/overview.png\" alt=\"NLNet\"\u003e\n\u003c/div\u003e\n\n(i) **Image Preprocessing**: The input image is pre-processed. 👉 Check [this file](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/lib/image_process.py#L17).\u003cbr\u003e\n(ii) **Graph Neural Network – Non-Local Modeling Method**: A two-stage GNN approach is presented for the non-local feature extraction and long-range dependency construction among different regions. The first stage aggregates local features inside superpixels. The following stage learns the non-local features and long-range dependencies among the graph nodes. It then integrates short- and long-range information based on an attention mechanism. The means and standard deviations of the non-local features are obtained from the graph feature signals. 👉 Check [this file](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/model/network.py#L62).\u003cbr\u003e\n(iii) **Pre-trained VGGNet-16 – Local Modeling Method**: Local feature means and standard deviations are derived from the pre-trained VGGNet-16 considering the hierarchical degradation process of the HVS. 👉 Check [this file](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/model/network.py#L37).\u003cbr\u003e\n(iv) **Feature Mean \u0026 Std Fusion and Quality Prediction**: The means and standard deviations of the local and non-local features are fused to deliver a robust and comprehensive representation for quality assessment. 👉 Check [this file](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/model/network.py). Besides, the distortion type identification loss $L_t$ , quality prediction loss $L_q$ , and quality ranking loss $L_r$ are utilized for training the NLNet. 👉 Check [this file](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/model/solver.py#L171). During inference, the final quality of the image is the averaged quality of all the non-overlapping patches. 👉 Check [this file](https://github.com/SuperBruceJia/NLNet-IQA/blob/main/lib/image_process.py#L17). \n\n### Poster Presentation\n\u003cdiv\u003e\n    \u003cdiv style=\"text-align:center\"\u003e\n    \u003cimg width=100%device-width src=\"https://github.com/SuperBruceJia/NLNet-IQA/raw/main/MMSP22_Poster.png\" alt=\"Poster\"\u003e\n\u003c/div\u003e\n\n## Structure of the Code\nAt the root of the project, you will see:\n```text\n├── main.py\n├── model\n│   ├── layers.py\n│   ├── network.py\n│   └── solver.py\n├── superpixel\n│   └── slic.py\n├── lib\n│   ├── image_process.py\n│   ├── make_index.py\n│   └── utils.py\n├── data_process\n│   ├── get_data.py\n│   └── load_data.py\n├── benchmark\n│   ├── CSIQ_datainfo.m\n│   ├── CSIQfullinfo.mat\n│   ├── KADID-10K.mat\n│   ├── LIVEfullinfo.mat\n│   ├── TID2013fullinfo.mat\n│   ├── database.py\n│   └── datainfo_maker.m\n├── save_model\n│   └── README.md\n├── test_images\n│   └── cr7.jpg\n├── real_testing.py\n```\n\n## Citation\nIf you find our work useful in your research, please consider citing it in your publications. \nWe provide a BibTeX entry below.\n\n```bibtex\n@inproceedings{Jia2022NLNet,\n    title     = {No-reference Image Quality Assessment via Non-local Dependency Modeling},   \n    author    = {Jia, Shuyue and Chen, Baoliang and Li, Dingquan and Wang, Shiqi},  \n    booktitle = {2022 IEEE 24th International Workshop on Multimedia Signal Processing (MMSP)},   \n    year      = {Sept. 2022},\n    volume    = {},\n    number    = {},\n    pages     = {01-06},\n    doi       = {10.1109/MMSP55362.2022.9950035}\n}\n\n@article{Jia2022NLNetThesis,\n    title     = {No-reference Image Quality Assessment via Non-local Modeling},\n    author    = {Jia, Shuyue},\n    journal   = {CityU Scholars},\n    year      = {May 2023},\n    publisher = {City University of Hong Kong},\n    url       = {https://scholars.cityu.edu.hk/en/theses/noreference-image-quality-assessment-via-nonlocal-modeling(2d1e72fb-2405-43df-aac9-4838b6da1875).html}\n}\n```\n\n## Contact\nIf you have any questions, please drop me an email at shuyuej@ieee.org.\n\n## Acknowledgement\nThe authors would like to thank Dr. Xuhao Jiang, Dr. Diqi Chen, and Dr. Jupo Ma for helpful discussions and invaluable inspiration. A special appreciation should be shown to Dr. Dingquan Li because this code is built upon his [(Wa)DIQaM-FR/NR](https://github.com/lidq92/WaDIQaM) re-implementation.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsuperbrucejia%2Fnlnet-iqa","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsuperbrucejia%2Fnlnet-iqa","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsuperbrucejia%2Fnlnet-iqa/lists"}