{"id":16172120,"url":"https://github.com/benehiko/tensorflowobjectdetectautolabel","last_synced_at":"2025-10-25T11:51:12.221Z","repository":{"id":139559303,"uuid":"168773157","full_name":"Benehiko/TensorflowObjectdetectAutoLabel","owner":"Benehiko","description":"Auto Label training data using a small trained model","archived":false,"fork":false,"pushed_at":"2019-02-02T00:24:19.000Z","size":18249,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-02-13T12:17:16.084Z","etag":null,"topics":["pascal-voc","tensorflow","tensorflow-object-detection-api"],"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/Benehiko.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":"2019-02-01T23:26:24.000Z","updated_at":"2022-05-18T02:47:02.000Z","dependencies_parsed_at":null,"dependency_job_id":"bbc18127-e437-48de-849b-77aef13c2973","html_url":"https://github.com/Benehiko/TensorflowObjectdetectAutoLabel","commit_stats":{"total_commits":7,"total_committers":1,"mean_commits":7.0,"dds":0.0,"last_synced_commit":"0d19d291906e71435510503237150b40a38f24a2"},"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Benehiko%2FTensorflowObjectdetectAutoLabel","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Benehiko%2FTensorflowObjectdetectAutoLabel/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Benehiko%2FTensorflowObjectdetectAutoLabel/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Benehiko%2FTensorflowObjectdetectAutoLabel/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Benehiko","download_url":"https://codeload.github.com/Benehiko/TensorflowObjectdetectAutoLabel/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247618843,"owners_count":20967861,"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":["pascal-voc","tensorflow","tensorflow-object-detection-api"],"created_at":"2024-10-10T03:44:53.095Z","updated_at":"2025-10-25T11:51:12.147Z","avatar_url":"https://github.com/Benehiko.png","language":"Python","readme":"# TensorflowObjectdetectAutoLabel\nAuto Label training data using a small trained model trained by using a small hand labelled dataset\n\n\n### Requirements\n\n- Python (3 or 2)\n- [Tensorflow](https://github.com/tensorflow/tensorflow)\n- [pascal_voc_writer](https://github.com/AndrewCarterUK/pascal-voc-writer)\n- numpy\n- Pillow\n\n### How to use it\n\nThis repository gives you all of the required dataset and trained model to test the supplied scripts with. The dataset supplied is a small sample set of [Stanford Ai car dataset](http://ai.stanford.edu/~jkrause/cars/car_dataset.html)\n\nTo use your own dataset and custom model follow the [Tensorflow Object Detection API](https://github.com/tensorflow/models/tree/master/research/object_detection) to train the model and hand label the images with [labelImg](https://github.com/tzutalin/labelImg) or any PascalVoc labelling tool. Train for a couple thousand steps to get an adequate detection confidence (80% +). I used 200 hand labelled images to train with.\n\nAfter you have acquired the frozen graph, add it to a folder and specify it inside the python scripts under MODEL_NAME and PATH_TO_CKPT. Also ensure to have the label map added here and specified under PATH_TO_LABELS. \n\nThe Dataset/Raw folder is a dump folder for images scrapped or gathered which you would like to be sorted using your trained model. \n\nRun object_detection_runner.py to sort your classes into folders under the Dataset/ path. \n\n    python3 object_detection_runner.py\n    \nThen run auto_label.py to label the sorted classes (specify the classes in CLASS_NAME list). Change the other constants as required. \n\n    python3 auto_label.py\n    \nAfter the run is complete, check the labelled images by going to [labelImg](https://github.com/tzutalin/labelImg) and selecting \"Open Dir\" and selecting \"Change Save Dir\" to the class folder under \"Dataset/\". This will load the generated annotations for you to verify on each image. \n\nPlease note that auto_label and object_detection_runner is not owned by me and a lot of the coding done was made by Google and [other open source contributors](https://github.com/bourdakos1/Custom-Object-Detection)\n\nI only added some extra features to generate PascalVoc xml files for each class detected.\n\n### Sources\n\n[Stanford Ai Car Dataset](http://ai.stanford.edu/~jkrause/cars/car_dataset.html)\n3D Object Representations for Fine-Grained Categorization\nJonathan Krause, Michael Stark, Jia Deng, Li Fei-Fei\n4th IEEE Workshop on 3D Representation and Recognition, at ICCV 2013 (3dRR-13). Sydney, Australia. Dec. 8, 2013.\n\n[Tensorflow](https://github.com/tensorflow/tensorflow)\n\n[Tensorflow Object Detection API](https://github.com/tensorflow/models/tree/master/research/object_detection)\n\n\n### License\n\n[Apache License 2.0](https://github.com/Benehiko/TensorflowObjectdetectAutoLabel/blob/master/LICENSE)\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbenehiko%2Ftensorflowobjectdetectautolabel","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbenehiko%2Ftensorflowobjectdetectautolabel","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbenehiko%2Ftensorflowobjectdetectautolabel/lists"}