{"id":24795996,"url":"https://github.com/licht-t/tf-centernet","last_synced_at":"2025-09-10T17:46:13.078Z","repository":{"id":57474793,"uuid":"297593277","full_name":"Licht-T/tf-centernet","owner":"Licht-T","description":"CenterNet implementation with Tensorflow 2","archived":false,"fork":false,"pushed_at":"2020-09-24T13:52:42.000Z","size":2156,"stargazers_count":7,"open_issues_count":1,"forks_count":1,"subscribers_count":3,"default_branch":"master","last_synced_at":"2024-12-16T11:56:20.735Z","etag":null,"topics":["centernet","object-detection","object-recognition","python","python3","tensorflow","tensorflow2"],"latest_commit_sha":null,"homepage":"","language":"Python","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/Licht-T.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":"2020-09-22T09:02:16.000Z","updated_at":"2024-11-27T05:07:50.000Z","dependencies_parsed_at":"2022-09-10T02:22:15.594Z","dependency_job_id":null,"html_url":"https://github.com/Licht-T/tf-centernet","commit_stats":null,"previous_names":[],"tags_count":7,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Licht-T%2Ftf-centernet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Licht-T%2Ftf-centernet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Licht-T%2Ftf-centernet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Licht-T%2Ftf-centernet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Licht-T","download_url":"https://codeload.github.com/Licht-T/tf-centernet/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":236279863,"owners_count":19123447,"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":["centernet","object-detection","object-recognition","python","python3","tensorflow","tensorflow2"],"created_at":"2025-01-30T00:30:28.480Z","updated_at":"2025-01-30T00:30:29.098Z","avatar_url":"https://github.com/Licht-T.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# tf-centernet\n[![PyPI version](https://badge.fury.io/py/tf-centernet.svg)](https://badge.fury.io/py/tf-centernet)\n![Upload Python Package](https://github.com/Licht-T/tf-centernet/workflows/Upload%20Python%20Package/badge.svg)\n\n[CenterNet](https://arxiv.org/abs/1904.07850) implementation with Tensorflow 2.\n\n## Install\n```bash\npip instal tf-centernet\n```\n\n## Example\n### Object detection\n```python\nimport numpy as np\nimport PIL.Image\nimport centernet\n\n# Default: num_classes=80\nobj = centernet.ObjectDetection(num_classes=80)\n\n# Default: weights_path=None\n# num_classes=80 and weights_path=None: Pre-trained COCO model will be loaded.\n# Otherwise: User-defined weight file will be loaded.\nobj.load_weights(weights_path=None)\n\nimg = np.array(PIL.Image.open('./data/sf.jpg'))[..., ::-1]\n\n# The image with predicted bounding-boxes is created if `debug=True`\nboxes, classes, scores = obj.predict(img, debug=True)\n```\n![output_obj](https://raw.githubusercontent.com/Licht-T/tf-centernet/master/data/output_obj.png)\n\n### Pose estimation\n```python\nimport numpy as np\nimport PIL.Image\nimport centernet\n\n# Default: num_joints=17\npe = centernet.PoseEstimation(num_joints=17)\n\n# Default: weights_path=None\n# num_joints=17 and weights_path=None: Pre-trained COCO model will be loaded.\n# Otherwise: User-defined weight file will be loaded.\npe.load_weights(weights_path=None)\n\n# Adjust this for the better prediction\npe.score_threshold = 0.1\n\nimg = np.array(PIL.Image.open('./data/chi.jpg'))[..., ::-1]\n\n# The image with predicted keypoints is created if `debug=True`\nboxes, keypoints, scores = pe.predict(img, debug=True)\n```\n![output_pose](https://raw.githubusercontent.com/Licht-T/tf-centernet/master/data/output_pose.png)\n\n\n## TODO\n* [x] Object detection\n* [x] Pre-trained model for object detection with Hourglass-104\n* [x] Pose estimation\n* [x] Pre-trained model for pose estimation with Hourglass-104\n* [ ] DLA-34 backbone and pre-trained models\n* [ ] Training function and Loss definition\n* [ ] Training data augmentation\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flicht-t%2Ftf-centernet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flicht-t%2Ftf-centernet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flicht-t%2Ftf-centernet/lists"}