{"data":{"slug":"oaid-tenginekit","name":"TengineKit","tagline":"TengineKit - Real-Time Mobile AI Algorithm SDK","github_url":"https://github.com/OAID/TengineKit","owner":"OAID","repo":"TengineKit","owner_avatar_url":"https://avatars.githubusercontent.com/u/29125077?v=4","primary_language":"C++","stars":2320,"forks":307,"topics":["ai","android","artificial-intelligence","computer-vision","deep-neural-networks","face-alignment","face-api","face-attributes","face-detection","face-landmarks","face-tracking","facial-landmarks","java","mobile","pytorch","tensorflow"],"archived":false,"github_pushed_at":"2021-10-18T07:27:43+00:00","maintenance_label":"Dormant","url":"https://www.graphcanon.com/tools/oaid-tenginekit","markdown_url":"https://www.graphcanon.com/tools/oaid-tenginekit.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/oaid-tenginekit","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=oaid-tenginekit","description":"TengineKit - Free, Fast, Easy, Real-Time Face Detection & Face Landmarks & Face Attributes & Hand Detection & Hand Landmarks & Body Detection & Body Landmarks & Iris Landmarks & Yolov5 SDK On Mobile.","homepage_url":null,"license":"Other","open_issues":32,"watchers":63,"ai_summary":"A comprehensive library for real-time detection and recognition tasks targeting mobile platforms like face and body landmarks, attributes, hand detections.","readme_excerpt":"[中文版本](Docs/README_CN.md)\n\n\n=======================================================================\n\n \n\nTengineKit, developed by OPEN AI LAB. \nTengineKit is an easy-to-integrate AI algorithm SDK. At present, it can run on various mobile phones at very low latency.**We will continue to update this project for better results and better performance!**\n\n# Effect\n\n| Face Detection &
Face 2dLandmark | Face 3dLandmark &
Iris | Upper Body Detection &
Uppper Body Landmark | Hand Detection &
Hand Landmark |\n| :---: | :---: | :---: | :---: |\n|
|
|
|
|\n\n\n\n## Gif\n
\n
dance of host
\n\n## Video( YouTube | BiliBili )\n[
](https://youtu.be/bnyD3laX_bU)\n
\n\n# Have a try\n- [Apk](Android/apk/TengineKitDemo-v1.0.3.apk) can be directly downloaded and installed on the phone to see the effect.\n\nor\n\n- scan code to download apk \n\n\n\n# Goals\n- Provide best performance in mobile client\n- Provide the simplest API in mobile client\n- Provide the smallest package in mobile client\n\n# Features\n- face detection\n- face landmarks\n- face 3dlandmarks\n- face attributes for example: age, gender, smile, glasses\n- eye iris & landmarks\n- body detect\n- hand detect(Real-time, not yet on Mobile)\n- hand landmarks(Real-time, not yet on Mobile)\n- body detect google(Real-time, not yet on Mobile)\n- body landamrks(Real-time, not yet on Mobile)\n- yolov5\n\n# Update (2021/03/25)\n- Fixed Linux sample code errer\n- Update Android sample code, up fps\n- update Linux so file\n- update Linux yolov5s\n- Fixed memory(Core v0.0.6)\n\n# Performance(Face Detect & Face Landmark)\n\n| CPU | Time consuming | Frame rate |\n| :---: | :---: | :---: |\n| Kirin 980 | 4ms | 250fps | \n| Qualcomm 855 | 5ms | 200fps |\n| Kirin 970 | 7ms | 142fps |\n| Qualcomm 835 | 8ms | 125fps |\n| Kirin 710F| 9ms | 111fps |\n| Qualcomm 439 | 16ms | 62fps |\n| MediaTek Helio P60 | 17ms | 59fps |\n| Qualcomm 450B | 18ms | 56fps |\n\n# Landmark Points Order\n[Landmark Points Order](Docs/POINTORDER.md)\n\n# Contact\nAbout the use of TengineKit and face-related technical exchanges, you can join the following QQ groups(Group Answer:TengineKit):\n- TengineKit communication QQ group: 630836519\n- Scan to join group\n \n ","github_created_at":"2020-06-29T07:28:46+00:00","created_at":"2026-07-11T12:24:01.771961+00:00","updated_at":"2026-07-31T06:00:44.297221+00:00","categories":[{"slug":"computer-vision","name":"Computer Vision","url":"https://www.graphcanon.com/categories/computer-vision","markdown_url":"https://www.graphcanon.com/categories/computer-vision.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/computer-vision"}],"tags":[{"slug":"ai","name":"ai"},{"slug":"android","name":"android"},{"slug":"artificial-intelligence","name":"artificial-intelligence"},{"slug":"computer-vision","name":"computer-vision"},{"slug":"deep-neural-networks","name":"deep-neural-networks"},{"slug":"face-alignment","name":"face-alignment"},{"slug":"face-api","name":"face-api"},{"slug":"face-attributes","name":"face-attributes"}],"trust":{"provenance":{"is_fork":false,"github_id":275761687,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-07-31T06:00:43.587Z","maintenance":{"label":"Dormant","score":18,"methodology":"github_public_v1","releases_90d":0,"days_since_push":1746,"last_release_at":"2021-10-18T07:27:43Z"},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T12:24:15.022Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-07-31T06:00:44.042Z"},"languages":{"value":["c++"],"source":"github.language","observed_at":"2026-07-31T06:00:44.042Z"},"license_spdx":{"value":"Other","source":"github.license","observed_at":"2026-07-31T06:00:44.042Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":{"notes":["The tool requires C++ skills to be effectively utilized in development.","Performance optimization is reported on select mobile platforms, mainly focusing on Kirin and Qualcomm series."],"min_ram_gb":null,"requires_docker":false},"constraints":{"min_ram_gb":null,"requires_docker":false},"when_to_use":["You aim to integrate comprehensive real-time AI algorithms on mobile devices such as face detection, landmarks, and attributes with low latency.","Specific device support is required for Kirin or Qualcomm processors where TengineKit shows optimized performance rates."],"when_not_to_use":["Your project does not require features like hand or body detection/landmarks in real-time on mobile devices, as these are not fully supported yet on the mobile platform provided by this SDK.","The necessity of using an open-source license is a critical factor for your decision since TengineKit's license is listed as 'Other', suggesting restrictions may apply."],"source":"enrich:decision_facts","observed_at":"2026-07-17T03:14:40.623Z"},"constraint_facets":{"min_ram_gb":null,"requires_docker":false},"decision_summary":[{"label":"Requirements","value":"The tool requires C++ skills to be effectively utilized in development.; Performance optimization is reported on select mobile platforms, mainly focusing on Kirin and Qualcomm series."},{"label":"Adopt for","value":"TengineKit is an easy-to-integrate AI algorithm SDK targeting mobile platforms for real-time detection and recognition tasks like face and body landmarks, attributes, hand detections."}]}}