Closed sumit1212 closed 5 years ago
The detection time is highly dependent on your project setup on which you are running the project - hardware (CPU/GPU), network between server and the client, and what is the size of your video: https://github.com/btwardow/tf-face-recognition/blob/master/server/static/local.js#L7-L9
The capturing has also quite few parameters. All is happening here: https://github.com/btwardow/tf-face-recognition/blob/master/server/static/detect.js where you can play around with some parameters.
Only one thing to notice here is that if you grab all the examples really fast, there's a risk they will be too similar. This can influence the recognition performance as well as the number of taken examples.
Thank you for your response.
What would be apply to detect face fast and moving detection with faces.
Thanks
Please give me help for 3 points. what can I do to solve these issues.