{"id":14958930,"url":"https://github.com/averdones/star_wars_object_detection","last_synced_at":"2025-10-24T16:30:39.577Z","repository":{"id":139446667,"uuid":"107145546","full_name":"averdones/star_wars_object_detection","owner":"averdones","description":"Using Tensorflow's Object Detection API to detect R2-D2 and BB-8 from Star Wars in videos and images.","archived":false,"fork":false,"pushed_at":"2020-03-26T13:04:14.000Z","size":48528,"stargazers_count":15,"open_issues_count":0,"forks_count":1,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-01-31T02:43:29.394Z","etag":null,"topics":["cnn","dataset","object-detection","python","star-wars","tensorflow-experiments"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","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/averdones.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":"2017-10-16T15:20:05.000Z","updated_at":"2024-07-26T20:23:21.000Z","dependencies_parsed_at":null,"dependency_job_id":"747c2420-ead5-4f02-bac5-06343cf00381","html_url":"https://github.com/averdones/star_wars_object_detection","commit_stats":{"total_commits":11,"total_committers":1,"mean_commits":11.0,"dds":0.0,"last_synced_commit":"7479a3ead96f5528e9af8e7d3e6fd8fc73012655"},"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/averdones%2Fstar_wars_object_detection","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/averdones%2Fstar_wars_object_detection/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/averdones%2Fstar_wars_object_detection/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/averdones%2Fstar_wars_object_detection/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/averdones","download_url":"https://codeload.github.com/averdones/star_wars_object_detection/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":237999466,"owners_count":19399886,"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":["cnn","dataset","object-detection","python","star-wars","tensorflow-experiments"],"created_at":"2024-09-24T13:18:33.068Z","updated_at":"2025-10-24T16:30:34.516Z","avatar_url":"https://github.com/averdones.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Star Wars Object Detection\nThis repository contains the dataset and some of the configuration files to train an object detector of R2-D2 and BB-8 from Star Wars, \nusing [Tensorflow's object detection API](https://github.com/tensorflow/models/tree/master/research/object_detection).\n\nThe final model has been fine-tuned using one of the [models pre-trained](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md)\non the [COCO dataset](http://mscoco.org/) as a starting point (*transfer learning*).\n\n_You can read more [here](https://averdones.github.io/tensorflow-object-detection-star-wars/)_.\n\n## Some results\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"results/result_1.gif\"\u003e\n\u003c/p\u003e\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"results/result_2.gif\"\u003e\n\u003c/p\u003e\n\n\u003ca href=\"https://youtu.be/tIYGObVra6E\" target=\"_blank\"\u003eHere there's a video to show the performance of the model.\u003c/a\u003e\n\nThe outcome seems acceptable considering the small size and low variability of the dataset used.\n\n## Trying the dataset\nIn order to try the dataset, you first need to follow the [installation instructions](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/installation.md) on the Tensorflow page. \nThen, you can train your own model following analogous commands as the ones used in the [pet detector](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/running_pets.md).\n\n### File description\n#### Folders\n* *annotations*: *xml* files with the boundig boxes of each image, obtained using [labelImg](https://github.com/tzutalin/labelImg), and a text file assigning each image to one of the two available classes (R2-D2 and BB-8). Analogous format as the one of the [Oxford-IIIT Pets Dataset](http://www.robots.ox.ac.uk/~vgg/data/pets/).\n* *graph_images*: plots of the Total Loss of the trained model.\n* *images*: 203 jpg train images, 102 of R2-D2 and 101 of BB-8.\n* *object_detection/data*: configuration file of the net trained, label map and tensorflow records (created with [create_sw_tf_record.py](create_sw_tf_record.py)).\n* *results*: some images/gifs showing the performance of the trained model.\n* *test_images*: some images not used to train the model.\n\n#### Scripts\n* *check_duplicate_images.py*: script that allows you to check if there are duplicate images in one or two directories. Useful when collecting the train and test images.\n* *create_sw_tf_record.py*: modified version of [create_pet_tf_record.py](https://github.com/tensorflow/models/blob/master/research/object_detection/create_pet_tf_record.py), where the paths have been changed along with the regular expression in *line 60* to adapt to my images names.\n* *export_inference_graph.py*: an exact copy of [export_inference_graph.py](https://github.com/tensorflow/models/blob/master/research/object_detection/export_inference_graph.py) in Tensorflow's API.\n* *object_detection_sw.ipynb*: ipython notebook to test the trained model on images and videos. Modified from [object_detection_tutorial.ipynb](https://github.com/tensorflow/models/blob/master/research/object_detection/object_detection_tutorial.ipynb) and some lines of code to test videos from [priya-dwivedi github](https://github.com/priya-dwivedi/Deep-Learning/blob/master/Object_Detection_Tensorflow_API.ipynb) (I don't know if that's the original source).\n* *rename_images.py*: script to rename all images files in one directory to have the same root name.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faverdones%2Fstar_wars_object_detection","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Faverdones%2Fstar_wars_object_detection","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faverdones%2Fstar_wars_object_detection/lists"}