{"id":17132805,"url":"https://github.com/araxeus/png-upscale","last_synced_at":"2025-04-10T03:57:02.354Z","repository":{"id":39634396,"uuid":"339779618","full_name":"Araxeus/PNG-Upscale","owner":"Araxeus","description":"AI Super - Resolution","archived":false,"fork":false,"pushed_at":"2023-06-30T12:17:55.000Z","size":162485,"stargazers_count":226,"open_issues_count":22,"forks_count":24,"subscribers_count":3,"default_branch":"main","last_synced_at":"2025-04-10T03:56:51.340Z","etag":null,"topics":["java","javacv","machine-learning","png","resampling","resize","resizing-images","software","super-resolution","upscaling"],"latest_commit_sha":null,"homepage":"","language":"Java","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/Araxeus.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}},"created_at":"2021-02-17T16:08:34.000Z","updated_at":"2025-03-31T18:30:55.000Z","dependencies_parsed_at":"2024-10-31T13:13:28.177Z","dependency_job_id":null,"html_url":"https://github.com/Araxeus/PNG-Upscale","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Araxeus%2FPNG-Upscale","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Araxeus%2FPNG-Upscale/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Araxeus%2FPNG-Upscale/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Araxeus%2FPNG-Upscale/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Araxeus","download_url":"https://codeload.github.com/Araxeus/PNG-Upscale/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248154997,"owners_count":21056542,"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":["java","javacv","machine-learning","png","resampling","resize","resizing-images","software","super-resolution","upscaling"],"created_at":"2024-10-14T19:28:24.051Z","updated_at":"2025-04-10T03:57:02.335Z","avatar_url":"https://github.com/Araxeus.png","language":"Java","funding_links":[],"categories":[],"sub_categories":[],"readme":"# :fire: PNG Upscale -  AI Super Resolution :fire:\n\u003e Small tool using pretrained models to upscale images\n \n ## Download is available from the [Releases Page](https://github.com/Araxeus/PNG-Upscale/releases/latest) or [Google Drive](https://drive.google.com/drive/folders/1pMvzL5sqTRcJOhByU_Am5vU2oik_KHH2?usp=sharing) or [MediaFire](https://app.mediafire.com/nz88rdbl8u041)\n* Hosted Folder include full \"[Models](https://github.com/Araxeus/PNG-Upscale/tree/main/Models)\" folder 📁 and executable Files 🖼️ to download\n\n* `Windows 64bit` [[Exe]](https://github.com/Araxeus/PNG-Upscale/releases/download/v1.0/png-upscale-1.0_Windows64bit.exe) / [[Jar]](https://github.com/Araxeus/PNG-Upscale/releases/download/v1.0/png-upscale-1.0_Windows64bit.jar) \n* `Linux 64bit` [[Jar]](https://github.com/Araxeus/PNG-Upscale/releases/download/v1.0/png-upscale-1.0_LINUX_64bit.jar) \n* `macOS 64bit` [[Jar]](https://github.com/Araxeus/PNG-Upscale/releases/download/v1.0/png-upscale-1.0_macOS_64bit.jar)\n* :electron:\tDownload the executable corresponding with your operating system, and the Models folder\n* It's possible to download only some of the models if you want (It just wont let you use them inside the program)\n* The Models folder needs to be in the same directory as the Jar/Exe to use them\n ---\n* ⚠️ Be careful when trying to resize very large pictures, it can take considerable time and resources ⚠️\n* To upscale an image you just need to choose a mode, load a picture and press start\n* Save button \u003cins\u003e*can*\u003c/ins\u003e be used to choose an output folder and filename *before* you start the process (either just name or .png)\n* You can double click the text box to change [Dark \u003c-\u003e Light] theme (disabled when upScaling)\n* Use PNG images for best results\n* if faced with a JNI Error see this issue for a possible fix https://github.com/Araxeus/PNG-Upscale/issues/33\n ---\n\n## Models\n\n\u003e`all of the model download links below are already included in the MediaFire folder.`\n\nThere are four trained models integrated into the program :\n\n### EDSR \n##### `[Best Quality]+[Slowest]`\n\n- Size of the model: ~38.5MB x3. This is a quantized version, so that it can be uploaded to GitHub. (Original was 150MB each.)\n- This model was trained for 3 days with a batch size of 16\n- Link to implementation code: https://github.com/Saafke/EDSR_Tensorflow\n- x2, x3, x4 trained models available\n- Advantage: \u003cins\u003e*Highly accurate*\u003c/ins\u003e\n- Disadvantage: \u003cins\u003e*Slow*\u003c/ins\u003e and large filesize\n- Speed: \u003c 3 sec for every scaling factor on 256x256 images on an Intel i7-9700K CPU.\n- Original paper: [Enhanced Deep Residual Networks for Single Image Super-Resolution](https://arxiv.org/pdf/1707.02921.pdf) [1]\n\n\u003e Trained models can be downloaded from [here](https://github.com/Saafke/EDSR_Tensorflow/tree/master/models).\n\n### ESPCN\n##### `[Fast]`\n\n- Size of the model: ~100kb x3\n- This model was trained for ~100 iterations with a batch size of 32\n- Link to implementation code: https://github.com/fannymonori/TF-ESPCN\n- x2, x3, x4 trained models available\n- Advantage: It is tiny and fast, and still performs well.\n- Disadvantage: Perform worse visually than newer, more robust models.\n- Speed: \u003c 0.01 sec for every scaling factor on 256x256 images on an Intel i7-9700K CPU.\n- Original paper: [Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network](https://arxiv.org/pdf/1707.02921.pdf) [2]\n\n\u003e Trained models can be downloaded from [here](https://github.com/fannymonori/TF-ESPCN/tree/master/export).\n\n### FSRCNN\n##### `[Fast]`\n\n- Size of the model: ~40KB x3\n- This model was trained for ~30 iterations with a batch size of 1\n- Link to implementation code: https://github.com/Saafke/FSRCNN_Tensorflow\n- Advantage: Fast, small and accurate\n- Disadvantage: Not state-of-the-art accuracy\n- Speed: \u003c 0.01 sec for every scaling factor on 256x256 images on an Intel i7-9700K CPU.\n- Notes: FSRCNN-small has fewer parameters, thus less accurate but faster.\n- Original paper: [Accelerating the Super-Resolution Convolutional Neural Network](http://mmlab.ie.cuhk.edu.hk/projects/FSRCNN.html) [3]\n\n\u003e Trained models can be downloaded from [here](https://github.com/Saafke/FSRCNN_Tensorflow/tree/master/models).\n\n### LapSRN\n##### `[Has x8]`\n\n- Size of the model: between 1-5Mb x3\n- This model was trained for ~50 iterations with a batch size of 32\n- Link to implementation code: https://github.com/fannymonori/TF-LAPSRN\n- x2, x4, x8 trained models available\n- Advantage: The model can do multi-scale super-resolution with one forward pass. It can now support 2x, 4x, 8x, and [2x, 4x] and [2x, 4x, 8x] super-resolution.\n- Disadvantage: It is slower than ESPCN and FSRCNN, and the accuracy is worse than EDSR.\n- Speed: \u003c 0.1 sec for every scaling factor on 256x256 images on an Intel i7-9700K CPU.\n- Original paper: [Deep laplacian pyramid networks for fast and accurate super-resolution](https://arxiv.org/pdf/1707.02921.pdf) [4]\n\n\u003e Trained models can be downloaded from [here](https://github.com/fannymonori/TF-LapSRN/tree/master/export).\n\n ---\n\n### Benchmarks\n\nComparing different algorithms. Scale x4 on monarch.png\n\n|               | Inference time in seconds (CPU)| PSNR | SSIM |\n| ------------- |:-------------------:| ---------:|--------:|\n| ESPCN            |0.01159   | 26.5471 | 0.88116 |\n| EDSR             |3.26758     |**29.2404**  |**0.92112**  |\n| FSRCNN           | 0.01298   | 26.5646 | 0.88064 |\n| LapSRN           |0.28257    |26.7330   |0.88622  |\n| Bicubic          |0.00031 |26.0635  |0.87537  |\n| Nearest neighbor |**0.00014** |23.5628  |0.81741  |\n| Lanczos          |0.00101  |25.9115  |0.87057  |\n\n ---\n \n### As a Demo this image was resized from 256x256 to 85x85, and then upscaled using this program\n\n![Original](https://github.com/Araxeus/PNG-Upscale/blob/main/test/original.png)\n\n\n## x2 Demo (85x85 -\u003e 170x170)\n\n|      Original             |  Bicubic Interpolation    |        EDSR               |\n| ------------------------- |------------------------- |------------------------- |\n![Original](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x2/original.png)   |  ![Bicubic](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x2/input(BicubicX2).png) |  ![EDSR](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x2/input(EDSRx2).png) |\n\n|         ESPCN             |       FSRCNN              |        LapSRN             |\n| ------------------------- | ------------------------- | ------------------------- |\n![ESPCN](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x2/input(ESPCNx2).png) | ![FSRCNN](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x2/input(FSRCNNx2).png) |  ![LapSRN](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x2/input(LapSRNx2).png) |\n\n\u003e Bicubic Interpolation is the standart resizing technique used by most editing tools like photoship etc..\n\n\n## x4 Demo (85x85 -\u003e 340x340)\n\n|      Original             |  Bicubic Interpolation    |        EDSR               |\n| ------------------------- | ------------------------- | ------------------------- |\n![Original](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x4/original.png)   |  ![Bicubic](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x4/input(BicubicX4).png)|  ![EDSR](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x4/input(EDSRx4).png)|\n\n \n|        ESPCN             |       FSRCNN              |        LapSRN             |\n| ------------------------- | ------------------------- | ------------------------- |\n![ESPCN](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x4/input(ESPCNx4).png)   |  ![FSRCNN](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x4/input(FSRCNNx4).png)|  ![LapSRN](https://github.com/Araxeus/PNG-Upscale/blob/main/test/x4/input(LapSRNx4).png)|\n\n### References\n[1] Bee Lim, Sanghyun Son, Heewon Kim, Seungjun Nah, and Kyoung Mu Lee, **\"Enhanced Deep Residual Networks for Single Image Super-Resolution\"**, \u003ci\u003e 2nd NTIRE: New Trends in Image Restoration and Enhancement workshop and challenge on image super-resolution in conjunction with **CVPR 2017**. \u003c/i\u003e [[PDF](http://openaccess.thecvf.com/content_cvpr_2017_workshops/w12/papers/Lim_Enhanced_Deep_Residual_CVPR_2017_paper.pdf)] [[arXiv](https://arxiv.org/abs/1707.02921)] [[Slide](https://cv.snu.ac.kr/research/EDSR/Presentation_v3(release).pptx)]\n\n[2] Shi, W., Caballero, J., Huszár, F., Totz, J., Aitken, A., Bishop, R., Rueckert, D. and Wang, Z., **\"Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network\"**, \u003ci\u003eProceedings of the IEEE conference on computer vision and pattern recognition\u003c/i\u003e **CVPR 2016**. [[PDF](http://openaccess.thecvf.com/content_cvpr_2016/papers/Shi_Real-Time_Single_Image_CVPR_2016_paper.pdf)] [[arXiv](https://arxiv.org/abs/1609.05158)]\n\n[3] Chao Dong, Chen Change Loy, Xiaoou Tang. **\"Accelerating the Super-Resolution Convolutional Neural Network\"**, \u003ci\u003e in Proceedings of European Conference on Computer Vision \u003c/i\u003e**ECCV 2016**. [[PDF](http://personal.ie.cuhk.edu.hk/~ccloy/files/eccv_2016_accelerating.pdf)]\n[[arXiv](https://arxiv.org/abs/1608.00367)] [[Project Page](http://mmlab.ie.cuhk.edu.hk/projects/FSRCNN.html)]\n\n[4] Lai, W. S., Huang, J. B., Ahuja, N., and Yang, M. H., **\"Deep laplacian pyramid networks for fast and accurate super-resolution\"**, \u003ci\u003e In Proceedings of the IEEE conference on computer vision and pattern recognition \u003c/i\u003e**CVPR 2017**. [[PDF](http://openaccess.thecvf.com/content_cvpr_2017/papers/Lai_Deep_Laplacian_Pyramid_CVPR_2017_paper.pdf)] [[arXiv](https://arxiv.org/abs/1710.01992)] [[Project Page](http://vllab.ucmerced.edu/wlai24/LapSRN/)]\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faraxeus%2Fpng-upscale","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Faraxeus%2Fpng-upscale","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faraxeus%2Fpng-upscale/lists"}