{"id":23177339,"url":"https://github.com/abdo-reda/soccerglanceprototype","last_synced_at":"2025-10-25T06:31:02.356Z","repository":{"id":167615208,"uuid":"580607665","full_name":"Abdo-reda/SoccerGlancePrototype","owner":"Abdo-reda","description":"Currently, this is a prototype for the Action Spotting Model for the SoccerNetV2 Dataset.","archived":false,"fork":false,"pushed_at":"2023-05-20T23:04:03.000Z","size":32204,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2024-12-18T06:33:54.085Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","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/Abdo-reda.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,"publiccode":null,"codemeta":null}},"created_at":"2022-12-21T01:26:41.000Z","updated_at":"2023-05-20T23:03:40.000Z","dependencies_parsed_at":"2023-06-07T21:45:37.350Z","dependency_job_id":null,"html_url":"https://github.com/Abdo-reda/SoccerGlancePrototype","commit_stats":null,"previous_names":["abdo-reda/soccernetprototype"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Abdo-reda%2FSoccerGlancePrototype","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Abdo-reda%2FSoccerGlancePrototype/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Abdo-reda%2FSoccerGlancePrototype/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Abdo-reda%2FSoccerGlancePrototype/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Abdo-reda","download_url":"https://codeload.github.com/Abdo-reda/SoccerGlancePrototype/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":238093098,"owners_count":19415165,"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":"2024-12-18T06:33:13.709Z","updated_at":"2025-10-25T06:30:56.711Z","avatar_url":"https://github.com/Abdo-reda.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n\n### Setup Enviornment \n* conda create -y -n PrototypeEnv python=3.9\n* conda activate PrototypeEnv\n* conda install -y cudnn=8.2\n* conda install -y cudatoolkit=11.3\n* conda install pytorch torchvision torchaudio pytorch-cuda=11.7 -c pytorch -c nvidia\n* pip3 install -r requirements.txt\n\n-------------- \n### Notes \u0026 Prerequistes\n\n* This project mainly depends that an \u003cb\u003e nginx server with an rtmp module \u003c/b\u003e is already setup to accept streams on rtmp://localhost:1935/live/ \n    * However the code could be modified to accept streams from whenever, the SOURCE will just need to be modified in the server.py file.\n    * Morover, The nginx.config that was used is included in the project directory for reference.\n\n--------------\n### Using the Prototype\n\n* To activate the enviornment \n    * `conda activate PrototypeEnv`\n* To run the model server \n    * `python server/server.py`\n    * You can then open http://localhost:5000/ and click on process stream button to start generating processing the video stream (this will simply execute the main.py script)   \n* To run an example of a client server \n    * `python client_server/client_server.py`\n\n--------------\n### Project Structure\n\n* \u003cb\u003e Main Directory \u003c/b\u003e\n    * `main.py` this is the main script that is responsible for creating and executing other python scripts/processes that will process the video stream.\n    * `configuration.py` is responsible for setting up parameters (a yaml file) that was used in the action spotting model (creating the features and generating actions).\n    * `offline_prototype.py` this is the old script that generates actions for a downloaded video.\n* \u003cb\u003e Server Directory \u003c/b\u003e Contains the main logic for the model server.\n    * `server.py` this is the main model server, it processes the video streams and communicates with the api.\n    * `api_service.py` this is a script that contains the most common calls/functions to the api, for example sending highlights or match is live and so on ..\n* \u003cb\u003e Client Server Directory \u003c/b\u003e\n    * `client_server.py` this an example for what a client application would look like, this was created to test that the clients can recieve the highlights.\n* \u003cb\u003e Chunk_Generation \u003c/b\u003e\n    * `transcribe_audio.py` this script is reponsible for transcribing the audio chunks.\n    * `extract_highlights.py`this script is reponsible for generating highlights from the transcribed audio chunks.\n    * `extract_actions.py` this script is responsbile for generating actions from the transcriptions.\n    * `capture_stream.py` this script was used to capture a video stream from interent/youtube using streamlink and store that data in temp folder.\n    * `generate_chunks.py` this script was used to generate video chunks from the data that was caputred by capture_stream.py.\n* \u003cb\u003e Build_Models [Archived] \u003c/b\u003e\n    * This contains different scripts that will reponsible for training and creating an instance of the model that was used for action spotting.\n* \u003cb\u003e Feature_Generation [Archived] \u003c/b\u003e\n    * `generate_features.py` this script is reponsbile for extracting features which will later be used to extract actions from the video stream.\n* \u003cb\u003e Output_Generation [Archived] \u003c/b\u003e\n    * `generate_output.py` this script is reponsible for generating actions from the extracted features from the video.\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fabdo-reda%2Fsoccerglanceprototype","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fabdo-reda%2Fsoccerglanceprototype","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fabdo-reda%2Fsoccerglanceprototype/lists"}