This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
Hello there! Well basically the second answer in this topic describes why trying to vectorize cuda memory is quite contra intuitive and probably not the best way to go.
However Thrust does have some cuda memory vectorization, described right here: http://stackoverflow.com/questions/17064096/thrustdevice-vector-in-constant-memory
There is more information on this in the Thrust documentation on the device vector for memory management.
OpenCV doesnt allow this, well it is more C++ not standard supplying functionality for that.
