{"id":20685516,"url":"https://github.com/yas-sim/objectron-3d-object-detection-openvino","last_synced_at":"2025-10-19T04:41:12.928Z","repository":{"id":185235183,"uuid":"275577486","full_name":"yas-sim/objectron-3d-object-detection-openvino","owner":"yas-sim","description":"Sample implementation of 3D object detection with Intel OpenVINO","archived":false,"fork":false,"pushed_at":"2020-06-29T23:25:31.000Z","size":131,"stargazers_count":15,"open_issues_count":3,"forks_count":5,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-03-29T15:34:53.843Z","etag":null,"topics":["3d-object-detection","deep-learning","intel","mediapipe","object-detection","objectron","openvino","python","sample"],"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/yas-sim.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}},"created_at":"2020-06-28T12:11:18.000Z","updated_at":"2022-10-31T10:25:51.000Z","dependencies_parsed_at":"2023-08-01T07:09:27.502Z","dependency_job_id":null,"html_url":"https://github.com/yas-sim/objectron-3d-object-detection-openvino","commit_stats":null,"previous_names":["yas-sim/objectron-3d-object-detection-openvino"],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yas-sim%2Fobjectron-3d-object-detection-openvino","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yas-sim%2Fobjectron-3d-object-detection-openvino/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yas-sim%2Fobjectron-3d-object-detection-openvino/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yas-sim%2Fobjectron-3d-object-detection-openvino/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/yas-sim","download_url":"https://codeload.github.com/yas-sim/objectron-3d-object-detection-openvino/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":250250385,"owners_count":21399622,"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":["3d-object-detection","deep-learning","intel","mediapipe","object-detection","objectron","openvino","python","sample"],"created_at":"2024-11-16T22:27:41.420Z","updated_at":"2025-10-19T04:41:07.885Z","avatar_url":"https://github.com/yas-sim.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\r\n# Overview\r\nThis is a sample implementaion of 3D object detection application using Intel(r) Distribution of OpenVINO(tm) Toolkit. The program uses `objectron` model from [Google *mediapipe* project](https://github.com/google/mediapipe) to detect the objects from the input images.  \r\nIn this project, I took the `objectron` model in TF SavedModel format from [PINTO model zoo](https://github.com/PINTO0309/PINTO_model_zoo) in the GitHub.  \r\n\r\nこれはIntel(r) Distribution of OpenVINO(tm) Toolkitを使用した3D物体検出アプリケーションの参考実装です。プログラムは[Google *mediapipe*プロジェクト](https://github.com/google/mediapipe)の`objectron`モデルを使用しています。  \r\nこのプロジェクトではPINTOさんの[PINTO model zoo](https://github.com/PINTO0309/PINTO_model_zoo)からTF SavedModelフォーマットのモデルデータを使用させて頂いています。  \r\n\r\n**Example of Detection result**  \r\n![output](resources/output.jpg)  \r\n![heatmap](resources/heatmap.png)\r\n\r\n## 1. Prerequisites\r\n* [Intel Distribution of OpenVINO toolkit 2020.2 (or newer)](https://software.intel.com/content/www/us/en/develop/tools/openvino-toolkit.html)\r\n\r\n## 2. Download Objectron TensorFlow model from PINTO model zoo\r\n*Linux compatible environment is required to run `download.sh` script to get the model*\r\n### 2.1 Download the model download script\r\n Download `download.sh` from [https://github.com/PINTO0309/PINTO_model_zoo/tree/master/36_Objectron/01_float32](https://github.com/PINTO0309/PINTO_model_zoo/tree/master/36_Objectron/01_float32)\r\n### 2.2 Run `download.sh` to get `saved_model_object_detection_3d_chair/*`  \r\n\r\n```sh\r\n ./download.sh\r\n```\r\nYou'll get following files (and the other model files too).\r\n```sh\r\nsaved_model_object_detection_3d_chair/\r\n+ variables/\r\n| + variables.data-00000-of-00001\r\n| + variables.index\r\n+ saved_model.pb\r\n```\r\n\r\n## 3. Convert objectron TF model into OpenVINO IR model\r\n\r\nUse `Model Optimizer (MO)` to convert the TF model into IR model.\r\n\r\nLinux\r\n```sh\r\npython3 ${INTEL_OPENVINO_DIR}/deployment_tools/model_optimizer/mo.py \\\r\n --saved_model_dir saved_model_object_detection_3d_chair \\\r\n --scale 255 \\\r\n --output_dir objectron_chair \\\r\n --data_type FP16\r\n```\r\nWindows\r\n```sh\r\npython \"%INTEL_OPENVINO_DIR%\\deployment_tools\\model_optimizer\\mo.py\" ^\r\n --saved_model_dir saved_model_object_detection_3d_chair ^\r\n --scale 255 ^\r\n --output_dir objectron_chair ^\r\n --data_type FP16\r\n```\r\nAn IR model `./objectron_chair/saved_model.xml|.bin` will be generated.\r\n\r\n## 4. Run sample program\r\n\r\n```sh\r\n(Linux) python3 objectron.py [-m model.xml] [-i input_image] [--heatmap]\r\n(Win10) python objectron.py [-m model.xml] [-i input_image] [--heatmap]\r\n```\r\n*Options*  \r\n`-m`: IR model path (default: `./objectron_chair/saved_model.xml`)  \r\n`-i`: Input image path. A webCam will be used as the input when you specify `cam` (default: `chair.jpg`)  \r\n`--heatmap`: Display heatmap\r\n\r\n**Command line example:**\r\n```sh\r\n $ python3 objectron.py -m ./objectron_chair/object_detection_3d_chair.xml -i cam --heatmap\r\n```\r\n\r\n## 5. Test Environment\r\n- Ubuntu 18.04 / Windows 10 1909  \r\n- OpenVINO 2020.3 LTS  \r\n\r\n## See Also  \r\n* [Google AI Blog, Real-Time 3D Object Detection on Mobile Devices with MediaPipe](https://ai.googleblog.com/2020/03/real-time-3d-object-detection-on-mobile.html)\r\n* [GitHub, Google *mediapipe* project](https://github.com/google/mediapipe)  \r\n* [GitHub, PINTO model zoo](https://github.com/PINTO0309)  \r\n* [Intel(r) Distribution of OpenVINO(tm) Toolkit](https://software.intel.com/content/www/us/en/develop/tools/openvino-toolkit.html)\r\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fyas-sim%2Fobjectron-3d-object-detection-openvino","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fyas-sim%2Fobjectron-3d-object-detection-openvino","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fyas-sim%2Fobjectron-3d-object-detection-openvino/lists"}