{"id":13543210,"url":"https://github.com/liuheng92/tensorflow_PSENet","last_synced_at":"2025-04-02T12:31:51.665Z","repository":{"id":49759130,"uuid":"174074825","full_name":"liuheng92/tensorflow_PSENet","owner":"liuheng92","description":"This is a tensorflow re-implementation of PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network.My blog:","archived":false,"fork":false,"pushed_at":"2020-02-21T09:24:56.000Z","size":1265,"stargazers_count":489,"open_issues_count":14,"forks_count":162,"subscribers_count":32,"default_branch":"master","last_synced_at":"2024-11-03T10:32:07.598Z","etag":null,"topics":["cpp","ocr","psenet","python","tensorflow","text-detection"],"latest_commit_sha":null,"homepage":"https://blog.csdn.net/liuxiaoheng1992/article/details/87646951","language":"C++","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/liuheng92.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}},"created_at":"2019-03-06T05:01:48.000Z","updated_at":"2024-10-03T03:36:19.000Z","dependencies_parsed_at":"2022-09-10T15:00:19.020Z","dependency_job_id":null,"html_url":"https://github.com/liuheng92/tensorflow_PSENet","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liuheng92%2Ftensorflow_PSENet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liuheng92%2Ftensorflow_PSENet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liuheng92%2Ftensorflow_PSENet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liuheng92%2Ftensorflow_PSENet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/liuheng92","download_url":"https://codeload.github.com/liuheng92/tensorflow_PSENet/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246815724,"owners_count":20838493,"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":["cpp","ocr","psenet","python","tensorflow","text-detection"],"created_at":"2024-08-01T11:00:26.146Z","updated_at":"2025-04-02T12:31:46.656Z","avatar_url":"https://github.com/liuheng92.png","language":"C++","funding_links":[],"categories":["Text detection and localization"],"sub_categories":["PSENet [paper:2018](https://arxiv.org/pdf/1806.02559.pdf)"],"readme":"# PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network\n\n### Introduction\nThis is a tensorflow re-implementation of [PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network](https://arxiv.org/abs/1806.02559).\n\nThanks for the author's ([@whai362](https://github.com/whai362)) awesome work!\n\n### Installation\n1. Any version of tensorflow version \u003e 1.0 should be ok.\n2. python 2 or 3 will be ok.\n\n### Download\ntrained on ICDAR 2015 (training set) + ICDAR2017 MLT (training set): \n\n[baiduyun](https://pan.baidu.com/s/14tQHf9MjuD0lSmwkoZhnCg)  extract code: pffd\n\n[google drive](https://drive.google.com/file/d/1TjJvtwMp8hJXQhn6Yz2lbPdvBGH-ZQ8u/view?usp=sharing)\n\nThis model is not as good as article's, it's just a reference.\nYou can finetune on it or you can do a lot of optimization based on this code. \n\n| Database | Precision (%) | Recall (%) | F-measure (%) | \n| - | - | - | - |\n| ICDAR 2015(val) | 74.61 | 80.93 | 77.64 |\n\n\n### Train\nIf you want to train the model, you should provide the dataset path, in the dataset path, a separate gt text file should be provided for each image, and **make sure that gt text and image file have the same names**.\n\nThen run train.py like:\n\n```\npython train.py --gpu_list=0 --input_size=512 --batch_size_per_gpu=8 --checkpoint_path=./resnet_v1_50/ \\\n--training_data_path=./data/ocr/icdar2015/\n```\n\nIf you have more than one gpu, you can pass gpu ids to gpu_list(like --gpu_list=0,1,2,3)\n\n**Note:**\n1. right now , only support icdar2017 data format input, like (116,1179,206,1179,206,1207,116,1207,\"###\"),\nbut you can modify data_provider.py to support polygon format input\n2. Already support polygon shrink by using pyclipper module\n3. this re-implementation is just for fun, but I'll continue to improve this code.\n4. re-implementation pse algorithm by using c++\n***(if you use python2, just run it, if python3, please replace python-config with python3-config in makefile)***\n\n### Test\nrun eval.py like:\n```\npython eval.py --test_data_path=./tmp/images/ --gpu_list=0 --checkpoint_path=./resnet_v1_50/ \\\n--output_dir=./tmp/\n```\n\na text file and result image will be then written to the output path.\n\n### Examples\n![result0](figure/result0.jpg)\n![result1](figure/result1.jpg)\n![result2](figure/result2.jpg)\n![result3](figure/result3.jpg)\n![result4](figure/result4.jpg)\n![result5](figure/result5.jpg)\n\n### About issues\nIf you encounter any issue check issues first, or you can open a new issue.\n\n### Reference\n1. http://download.tensorflow.org/models/resnet_v1_50_2016_08_28.tar.gz\n2. https://github.com/CharlesShang/FastMaskRCNN\n3. https://github.com/whai362/PSENet/issues/15\n4. https://github.com/argman/EAST\n\n### Acknowledge\n[@rkshuai](https://github.com/rkshuai) found a bug about concat features in model.py.\n\n**If this repository helps you，please star it. Thanks.**\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fliuheng92%2Ftensorflow_PSENet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fliuheng92%2Ftensorflow_PSENet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fliuheng92%2Ftensorflow_PSENet/lists"}