{"id":13620374,"url":"https://github.com/shieman/VideoProcess","last_synced_at":"2025-04-14T19:31:57.405Z","repository":{"id":217074445,"uuid":"255957389","full_name":"shieman/VideoProcess","owner":"shieman","description":"Denoise, Colorize, Upscale Resolution, Video Stabilization, Increase FPS.","archived":false,"fork":false,"pushed_at":"2020-04-29T15:23:20.000Z","size":1081,"stargazers_count":28,"open_issues_count":0,"forks_count":7,"subscribers_count":1,"default_branch":"master","last_synced_at":"2024-11-08T06:41:37.315Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/shieman.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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}},"created_at":"2020-04-15T15:21:24.000Z","updated_at":"2024-08-10T20:44:10.000Z","dependencies_parsed_at":null,"dependency_job_id":"8edfe17f-bf30-44f5-a063-1a5de086a16c","html_url":"https://github.com/shieman/VideoProcess","commit_stats":null,"previous_names":["shieman/videoprocess"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/shieman%2FVideoProcess","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/shieman%2FVideoProcess/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/shieman%2FVideoProcess/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/shieman%2FVideoProcess/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/shieman","download_url":"https://codeload.github.com/shieman/VideoProcess/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248945981,"owners_count":21187421,"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":[],"created_at":"2024-08-01T21:00:55.098Z","updated_at":"2025-04-14T19:31:56.257Z","avatar_url":"https://github.com/shieman.png","language":"C++","funding_links":[],"categories":["C++"],"sub_categories":[],"readme":"# VideoProcess\nI needed video and image processings for a theater project. so i implemented this tools.\nTools to denoise, colorize, upscale resolution, video stabilization, increase FPS to videos.\nAll that thanks to Machine Learning.\n### Tools :\n```\nWhite Balance\nDenoise\nVideo Stabilization\nDAIN (extern tool)\nSuper Resolution\nColorization\n```\n\n# White Balance\nColor White Balance.\nfrom opencv xphoto contrib module.\n\n# Denoise\nPerforms image denoising using the Block-Matching and 3D-filtering algorithm.\nIt works on Black and White images...\nfrom opencv xphoto contrib module.\n\n# Video Stabilization\nPerform video stabilization.\nFrom opencv videostab contrib module. \n\n# DAIN\n\"Dain-App is a free app that let you take any form of media like movies, stop-motion, anime, cartoons,\nsprites, etc and interpolate new frames, generating a bigger frame-rate from the original file.\"\n\nThe link : https://grisk.itch.io/dain-app\n\nSee tutos to use it. I won't explain.\nCuda app, so Nvidia graphic card only.\n\n# Super Resolution\nThis module contains several learning-based algorithms for upscaling an image.\n### The models :\n1. EDSR\n    - x2, x3, x4 trained models available\n    - Advantage: Highly accurate\n    - Disadvantage: Slow and large filesize\n\n2. 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    \n3. FSRCNN\n    - Advantage: Fast, small and accurate\n    - Disadvantage: Not state-of-the-art accuracy\n    \n4. LapSRN\n    - x2, x4, x8 trained models available\n    - Advantage: The model can do multi-scale super-resolution with one forward pass.\n    - Disadvantage: It is slower than ESPCN and FSRCNN, and the accuracy is worse than EDSR.\n    \nfrom opencv dnn superres contrib module.\nI hacked a part of source to include in mine to use possibility of cuda dnn.\n\n# Colorization\nColorize Black and white (even colored) images and videos.\nIt uses machine learning model to predict colors.\n\n# How it works\nWindows Only. Tested on Win 10.\nAll the modules, except DAIN app, uses opencv 4.2 cuda 10.2.\nSo it's better to have a Nvidia graphic card to perform max.\nSome of the tools are quite slow, even on GPU. Using them on CPU should take a lot of time.\n\n- download repository\n- install Cuda 10.2 and Cudnn 7.6.5.\n- For the Windows users, i share the dll's from opencv :\n  Opencv downloads : http://docs.encima.fr/index.php/s/zJnnmqgzd949cJD\n  \n  the file opencv_videoio_ffmpeg420_64.old is a renamed opencv_videoio_ffmpeg420_64.dll file because i did not\n  manage the use of ffmpeg for videos. I did not find a good way to use it, do not know why so renamed in .old,\n  opencv does not use it...\n\n  These files should be near .exe files or on a place that is on windows path in env vars.\n\n- Some tools use models, place them in Models folder : http://docs.encima.fr/index.php/s/QN9QHbw8ef5bXEN\n\nAll tools work in the same way, you can find help with --help for each one.\nSome parameters have default values, check thanks to --help.\nEach tool can process many files in same execution. They pick files in --input dir and export results in --output dir.\nThey can process :\n    - .jpg, .png, .tif images\n    - .mov, .mp4 and .avi video files.\n      Videos are exported in .mov with avc1 codec.\n\n### example :\nSuperRes.exe --i=Input --o=Output --algo=espcn --scale=3\n\nMultiply x3 the resolution of images and videos in Input folder, using Espcn algo, and export results in Output folder.\n\n# Thanks\n```\nOpencv team - https://opencv.org/\nOpencv contrib module xphoto, superres_dnn, videostab teams - https://docs.opencv.org/4.2.0/modules.html\nGrisk Dain app - https://grisk.itch.io/dain-app\nDain github team - https://github.com/baowenbo/DAIN\n```\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fshieman%2FVideoProcess","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fshieman%2FVideoProcess","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fshieman%2FVideoProcess/lists"}