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 was thinking to extend Yarrar project, as it is Unity3D ready, in order to create pluggable modules for processing any camera's frame but mixing instance or category matching tasks, even maybe facial landmark detection with dlib or dest. For that, nanomsg or nng sounds like to most optimal way to manage the input flow from the camera with out memory leaks.
I am really interested by your point of view about such bundling of caffe or mixed image recognition frameworks on mobile devices :-)
Have a great day !
Cheers,
Richard
The text was updated successfully, but these errors were encountered:
Hi,
Hope you are all well !
Just wanted to shared a couple of suggestions to bundle your caffe-mobile project for cross-platform compilation:
- CMake related
- Camera Pipeline related
- Deep Learning for Mobile related
I was thinking to extend Yarrar project, as it is Unity3D ready, in order to create pluggable modules for processing any camera's frame but mixing instance or category matching tasks, even maybe facial landmark detection with dlib or dest. For that, nanomsg or nng sounds like to most optimal way to manage the input flow from the camera with out memory leaks.
I am really interested by your point of view about such bundling of caffe or mixed image recognition frameworks on mobile devices :-)
Have a great day !
Cheers,
Richard
The text was updated successfully, but these errors were encountered: