Face Detection and Recognition Using GoCV

A simple face detection and recognition program written in golang by making use of Haar feature based cascade classifier

Requirements for Running it on Mac OS

• Go lang - version go1.13.6 darwin/amd64 (https://golang.org/dl/)
• Docker for Mac or Docker Desktop (https://docs.docker.com/docker-for-mac/install)
• OpenCV 3 - brew install opencv3
• Go Package Dependencies
  • go get gocv.io/x/gocv
  • go get github.com/machinebox/go-sdk/facebox

Steps to Run

1. Clone the repo
2. Register in machine-box (https://machinebox.io)
3. Get Machine-Box Access Key, looks something like this MB_KEY=xxxxxxxxxxx....
4. Run MachineBox locally
   • export your machine-box key export MB_KEY="xxxxx...."
   • run machinebox using docker docker run -d -p 8080:8080 -e "MB_KEY=$MB_KEY" machinebox/facebox remove -d if you dont want machine-box container to run in detached mode
5. Now check machinebox by hitting http://localhost:8080
6. Upload your image for training under Post a file -> Try it now
7. cd to /your/directory/main.go and go run main.go in your terminal

FAQ

How to improve recognition accuracy There are few suggestions:

• Purely uses Haar Cascade clssifier default face xml, hence no control over the algorithm
• Upload more samples of your faces in different postures to machinebox's facebox for better recognition
• Try to test it under sufficient lighting environment where your face is clearly visible, that said avoid low-light and less grain or noise when using webcam

Useful links for references

1. https://github.com/opencv/opencv/tree/master/data/haarcascades - Haar cascade classifier xml files
2. https://gocv.io
3. https://machinebox.io