You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I suppose if you use the logic of my implementation, you would be able to change any of the components. As far as regions are concerned, I am only using a box on the forehead that is derived from the output of the face detector. The method in the paper you linked is much more sophisticated, as it involves detecting detailed facial landmarks and using them to define many regions.
Enabling the detection via IR would make this project quite useful as a baby monitor, e.g. to alert the parents of the onset of a sudden infant death syndrome.
The current methods is to use pulse meters attached to the foot of an infant.
A cheap IR camera + a rasperry pi could be a great and accessible alternative that doesn't disturb the baby as much as a probe attached to a foot.
Do you think your method would work on an camera feed with irPass filter, like discussed here:
https://arxiv.org/pdf/1902.05194.pdf ?
In your code you get the mean RGB value. So maybe using the mean IR as they mention "we compute the mean IR value on each region" could work too?
The text was updated successfully, but these errors were encountered: