Version V3.0(2018/08/16 update)

Introduce

The eGroup AI face engine has ability to integrate with your AP. The sample code demo how to use the face engine by java.

Enviroment

• Wintel Enviroment Deep Learning Architecture Support Edge Computing Solution

• Wintel Enviroment (minimum)

1. Software：Windows 7／10、C++ Redistributable 2015
2. Hardware：4th Generation Intel® Core™ i3 (SSE3)
3. Input Source：
   • Web Camera：
     • Resolution：360p ~ 1080p
   • IP Camera
     • Resolution：360p ~ 4K
     • Protocol：RTSP、RTMP
     • Format：H.264、H.265、M-JPEG
   • Video File
     • Protocol：RTSP、RTMP
     • Format：MP4、MKV、MOV、AVI、FLV
   • Image File
     • Format：JPG、PNG

Function

Retrieve

Start the FaceRecognize engine and the engine will start to recognize the source you input, you can control cli to retrieve frame or face to your folder.

Train

You can use list.txt that contain face's path name face's name to train the model.

Model Switch

Switch your new model, e.g model1 is in used and you train the new model2 ,you can switch model2 don't need to restart the FaceRecognize engine.

Model Merge

Merge your new model, e.g If you have model1 and model2, you can merge to model3

GetCacheResult

If you want to real-time recognize ,you can find newest 100 record in the cache.result.json

GetAllResult

If you want to get the all recognize result , you can find in the output.json

Resource

• Official website : https://www.egroup.com.tw

• Youtube :

1. Reference 2018 IoT Seminar FaceCheckIn【Case】
2. Reference 2017 Taiwan Future Tech FaceCheckIn【Case】
3. CMD Practice【CPU Demo】
4. FaceCheckIn(JAVA GUI)【CPU Demo】
5. Multiple WebCam Recognition with one PC【CPU Demo】
6. Detect & Train at the same time【CPU Demo】
7. Object Detection【GPU Demo】

• Github : https://github.com/eGroupTeam

【Contact Information】

Contact Person：Daniel

Email：egroup.daniel@gmail.com

Powered By eGroup AI