Privacy Mask

[ugrulas]

I'm developing a privacy feature for a video recording application where the camera captures faces. I plan to mask sensitive areas like the eyes and lips. I'm thinking of using the MediaPipe Face Mesh model for this task. Is it feasible to implement this using a computer vision SDK? How can I integrate this into my project?

[github-actions[bot]]

This automatically generated reply acts as a friendly reminder.

Answers to your questions will most often come from the community, from developers like yourself. You will, from time to time, find that Axis employees answers some of the questions, but this is not a guarantee. Think of the discussion forum as a complement to other support channels, not a replacement to any of them.

If your question remains unanswered for a period of time, please revisit it to see whether it can be improved by following the guidelines listed in Axis support guidelines.

[vivekatoffice]

Hi @ugrulas ,
Which Axis camera model do you have?

Not sure if MediaPipe Face Mesh model can be converted into TensorFlow Lite (tflite) using the TensorFlow Lite converter.
DLPU model conversion

Test your model

[ugrulas]

Hi @vivekatoffice ,

I am using P3255-LVE and I am able to run blazeface face detection and facelandmark model using computer vision sdk. The models are already in tflite format but I dont know how to do masking. I want achieve something similar to AXIS Live Privacy Shield.

[vivekatoffice]

In acap-computer-vision-sdk it is not possible as Axis APIs for direct overlay on RTSP stream is not there, you may need to build it in acap-native-sdk and use Axis AxOverlay API.

You can have some type of websocket connection for sending the value from CV to Native SDK. You can explore the overlay capability using Cairo documentation

[ugrulas]

Thank you for the guidance