{"id":13441977,"url":"https://github.com/sanechips-multimedia/syenet","last_synced_at":"2025-03-20T13:31:44.379Z","repository":{"id":188448658,"uuid":"678708484","full_name":"sanechips-multimedia/syenet","owner":"sanechips-multimedia","description":"SYENet: A Simple Yet Effective Network for Multiple Low-Level Vision Tasks with Real-Time Performance on Mobile Device, in ICCV 2023","archived":false,"fork":false,"pushed_at":"2023-12-18T09:11:03.000Z","size":48,"stargazers_count":42,"open_issues_count":4,"forks_count":4,"subscribers_count":3,"default_branch":"main","last_synced_at":"2024-08-01T03:38:34.288Z","etag":null,"topics":["computer-vision","deep-neural-networks"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/sanechips-multimedia.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}},"created_at":"2023-08-15T07:21:06.000Z","updated_at":"2024-06-19T17:01:18.000Z","dependencies_parsed_at":null,"dependency_job_id":"d4d831bc-fbfd-4c03-b585-d43ba3e47ab8","html_url":"https://github.com/sanechips-multimedia/syenet","commit_stats":null,"previous_names":["sanechips-multimedia/syenet"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sanechips-multimedia%2Fsyenet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sanechips-multimedia%2Fsyenet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sanechips-multimedia%2Fsyenet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sanechips-multimedia%2Fsyenet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sanechips-multimedia","download_url":"https://codeload.github.com/sanechips-multimedia/syenet/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":221768427,"owners_count":16877638,"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":["computer-vision","deep-neural-networks"],"created_at":"2024-07-31T03:01:40.266Z","updated_at":"2025-03-20T13:31:44.362Z","avatar_url":"https://github.com/sanechips-multimedia.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# SYENet\n\nThis repository contains the official implementation for ICCV 2023 paper:\n\n[SYENet: A Simple Yet Effective Network for Multiple Low-Level Vision Tasks with Real-Time Performance on Mobile Device](https://openaccess.thecvf.com/content/ICCV2023/papers/Gou_SYENet_A_Simple_Yet_Effective_Network_for_Multiple_Low-Level_Vision_ICCV_2023_paper.pdf)\n\nWeiran Gou, Ziyao Yi, Yan Xiang, Shaoqing Li, Zibin Liu, Dehui Kong, Ke Xu. [[arxiv]](https://arxiv.org/abs/2308.08137)\n\nSYENet is an efficient network that could handle multiple low-level vision (isp, lle and sr) tasks. SYENet utilises re-parameterization for fast inference and got the highest score in [MAI 2022 Learned Smartphone Challenge](https://arxiv.org/abs/2211.03885).\n\n### Citation\n\nIf you find our work useful in your research, please cite:\n\n```\n@InProceedings{Gou_2023_ICCV,\n    author    = {Gou, Weiran and Yi, Ziyao and Xiang, Yan and Li, Shaoqing and Liu, Zibin and Kong, Dehui and Xu, Ke},\n    title     = {SYENet: A Simple Yet Effective Network for Multiple Low-Level Vision Tasks with Real-Time Performance on Mobile Device},\n    booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},\n    month     = {October},\n    year      = {2023},\n    pages     = {12182-12195}\n}\n```\n\n### Environment\n\n- python 3.8\n- pytorch == 1.12.1\n- numpy == 1.23.3\n- cv2 == 4.7.0\n- PIL == 9.2.0\n- tqdm == 4.64.1\n- yaml == 6.0\n\n### Configuration\n\nEdit the yaml files (isp.yaml, lle.yaml, sr.yaml) in ./config.\n\nYou are recommended to use [basicsr](https://github.com/XPixelGroup/BasicSR) to train our sr models to get higher PSNR. We put the train/test configuration files for training/testing our sr models using basicsr in ./config, which are sr_basicsr_train.yaml and sr_basicsr_test.yaml.\n\n### Train\n\nIf you want to re-parameterize the model and save it, please set 'train $\\rightarrow$ save_slim' parameter in the configuration yaml file to be true. And hence, the re-parameterized small model for fast inference will be saved.\n\nFor isp and lle tasks, we utilise a warmup phase which is a self-supervised training stage. This phase could be cancalled by setting 'train $\\rightarrow$ warmup' parameter in the configuration yaml file to be false.\n\n```bash\npython main.py -task train -model_type original -model_task isp/lle/sr -device cuda\n```\n\n### Test\n\nSet the 'model $\\rightarrow$ type' parameter in the configuration yaml file to be original if you are loading an original pretrained model, otherwise, set it to be re-parameterized for loading a re-parameterized model.\n\nIf you are loading an original pretrained model, but you want to re-parameterize it before inference, you could set 'model $\\rightarrow$ need_slim' parameter in the configuration yaml file to be true. Notice that you cannot re-parameterize a re-parameterized model.\n\nYou could save the images generated in the test by setting 'test $\\rightarrow$ save' parameter in the configuration yaml file to be true.\n\n```bash\npython main.py -task test -model_type original -model_task isp/lle/sr -device cuda\n```\n\n### Demo\n\nYou could save the images generated in the demonstration by setting 'demo $\\rightarrow$ save' parameter in the configuration yaml file to be true.\n\n```bash\npython main.py -task demo -model_type original -model_task isp/lle/sr -device cuda\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsanechips-multimedia%2Fsyenet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsanechips-multimedia%2Fsyenet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsanechips-multimedia%2Fsyenet/lists"}