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
{{ message }}
This repository has been archived by the owner on Dec 20, 2020. It is now read-only.
We should take advantage of GPU optimizations since we are using a jetson, and also most of us have CUDA installed on our machines.
OpenCV has cuda implementations for a majority of the computationally expensive functions, a lot of which we are using in our computer vision node. We should start porting these over and always look for a cuda implementation first to see if it's available.
We should take advantage of GPU optimizations since we are using a jetson, and also most of us have CUDA installed on our machines.
OpenCV has cuda implementations for a majority of the computationally expensive functions, a lot of which we are using in our computer vision node. We should start porting these over and always look for a cuda implementation first to see if it's available.
https://docs.opencv.org/4.1.0/d1/d1a/namespacecv_1_1cuda.html
This should apply to all aspects of the computer vision node: endline, traffic light, traffic sign, and lane detection
The text was updated successfully, but these errors were encountered: