The code of InsightFace Python Library is released under the MIT License. There is no limitation for both academic and commercial usage.
The pretrained models we provided with this library are available for non-commercial research purposes only, including both auto-downloading models and manual-downloading models.
- Front_end
- pyqt5 ->Windows desktop app for gate access control
- django ->web app for guys data upload ,management and analysis
- Back_end
- database
- milvus_lite ->face_embedding
- sqlite3 ->person_info
- algorithm
- general functions
- insightface ->face_recognition π©π»βππ¨π»βπ
- yolo v8/Nas ->car_board_recognition π
- database
- insightface
- milvus
- sqlite3
- flask
- test the new milvus server β
- refresh start
- no refresh start
- insert entries while searching
- sqlite connect with metabase β ->docker_starter
- flask log page
- sqlite design
- delete the useless process of start milvus server β
- try to make the milvus dynamic which means the server can be insert and delete dynamically β
- add func check entries before insert data into milvus β
- maybe the way of reading data from files can be optimized by asynchronous reading β³
- simplify the code of class Image β³
- try to use logging to record the process of program β³
- try to add method of test_videos in class FaceAnalysisTest β³
- make the init() of class Milvus,FaceAnalysis,--Test,Image more slightly β
- add the Error handling of class Milvus,FaceAnalysis,--Test,Image β³
- test insert ,_create_collection,_base_config_set methods in class Milvus β
- finished the construction of class Milvus β
- try to create collection and insert data into milvus_lite from npy files β
- try to use milvus_lite to search the face data β
- maybe the way of reading data from files can be optimized by asynchronous reading β³
- test insert ,_create_collection,_base_config_set methods in class Milvus β³
- set index parameters in milvus_lite β³
- finish the construction of class Milvus β³
- try to create collection and insert data into milvus_lite from npy files β³
- try to use milvus_lite to search the face data β³
- finished swap the face embedding data from npz to npy β
- completed construction of the function of insert,_create_collection,_base_config_set in class Milvus β
- finish the construction of class Milvus β³
- try to create collection and insert data into milvus_lite from npy files β³
- try to use milvus_lite to search the face data β³
- try to swap the face embedding data from npz to npy β³
- load images from mess folder with Path.rglob() β
- accelerate the process of get image's ndarray or face embedding by using npio β
- try a large register of face data which is 10000+ and prepared for milvus search β
- can Django interact with milvus_lite ?
- finish the construction of class Milvus β³
- almost figure out the API of Milvus β
- get more models which have the best accuracy of recognize south Asian from insightface and try to use them β
- move models folder out from project folder for easier to update git β
- stop using docker for pymilvus, use milvus (milvus lite) instead β
- docker is too complex to use
- milvus lite is easy to use directly
- got an example of milvus lite and start to figure out how to use it β
- server need to be stopped (or it'll fail to w/r next time) and restarted after each time of using it
- class Milvus in milvus_lite.py is constructing , finished the part of init() β
- docker install β
- error
- unexpected error was encountered while executing a WSL command
- solved by
- wsl -update
- error
- milvus install β
- image from docker hub
- milvusdb/milvus:latest
- python sdk
- pip install pymilvus->milvus
- pip install milvus->milvus lite
- image from docker hub