We implemented real-time face-tracking technology in this demo project. This project can be built on Ubuntu 22.04 and model inference was implemented for web camera and real-time video streaming
face_track_.mp4
- Build OpenCV on your platform, follow the steps in the link below. https://gist.github.com/kleysonr/c0752306bb6c021a1ff3c448996636ee
- Install vulkan sdk with the command "./install.sh"
Go to the project folder and enter the following command. ./build.sh
Go to the build folder and run with the below command
./<executive file name> --frame_num <frame number>
for example:
./face_track --frame_num 150
If you need to save video file with tracking, you can run the command below
./<executive file name> --video_out true
for example:
./face_track --video_out true
TrackData contains one cropped face image per person, this can be used later for recognition purpose ```shell TrackData track_data = GetTrackData(frame, tracks);
## Improvement
This project can be improved by using kby-ai's face recognition SDK. The SDK requires a license.
- To request a license, please contact us:
Email: contact@kby-ai.com
Telegram: @kbyai
WhatsApp: +19092802609
Skype: live:.cid.66e2522354b1049b