Is there any information or sample or tutorial on how to use the OpenCV HAL to speed up my code?
I noted that OpenCV 3 RC1 has a library called HAL which is designed to speed up the basic functions of OpenCV.
Can I use this library to speed up my code?
Is there any document / tutorial / ... on how to use this library?
I want to speed up my code so that it can run quickly on Intel and ARM processors.
source to share
If I understand this well, then HAL is just a handy, IPP-like, low-level API to speed up OpenCV across platforms (by hiding low-level operations, i.e. core, imgproc, ... => HAL) and to enable hardware suppliers for the implementation of accelerated visualization and vision algorithms.
OpenCV already had some acceleration API (i.e. parallel framework or OpenCL) and HAL fits into this sequence as new, HAL focuses on the inclusion of mobile and embedded systems and can be used directly by high-level libraries or applications.
The feature set can be found in the namespace cv::hal
- check out the HAL module documentation for the features implemented.
Obviously, you don't have to worry about these functions, because they are just (low-level) API internals; in OpenCV 3.0 they are also mentioned in some functions, for example. cv::magnitude()
evokes cv::hal::magnitude()
within itself. That is, if OpenCV was built with NEON support, then you will implicitly use the HAL.
source to share