Is it possible to manually map face landmarks for images where mediapipe doesn't detect them at all?

[aesanchezgh]

Hi all,

I have some difficult images where mediapipe is not detecting the face landmarks. If I lower the thresholds and increase the image brightness, it detects landmarks, but they are not accurate. Is there any method where I could use python code in some way to add all 478 points to an existing image I have?

Thanks!

[Ashlin-Gov]

I have been experiencing a similar issue with the face detection.

Any idea when the full range models will be released?

[aesanchezgh]

Any update on this issue?