GuidedFilter | Implementation of the Guided Image Filtering algorithm | Computer Vision library

GuidedFilter is a C++ library typically used in Artificial Intelligence, Computer Vision, OpenCV, Example Codes applications. GuidedFilter has no bugs, it has no vulnerabilities, it has a Permissive License and it has low support. You can download it from GitHub.

GuidedFilter is an implementation of the [Guided Image Filtering] algorithm in OpenCL. The Guided Filter is an image filter with many applications, one of which is edge-preserving smoothing. It has a non-approximate algorithm which is O(1) in the filter window size. It has excellent performance characteristics which make it a great alternative to the popular Bilateral Filter. You can watch how the algorithm performs as a smoothing operator in the following two videos: * [Guided Image Filtering on Kinect RGB stream with OpenCL] * [Guided Image Filtering on Kinect RGB and Depth streams with OpenCL] In the latter video, a point cloud is built from the Kinect streams. The Guided Image Filtering algorithm is applied to the Depth frame and each of the 3 channels of the RGB frame, separately. The frames are transfered to the GPU, processed with OpenCL, and then delivered directly to OpenGL. On my machine, I was able to get a mean running time of 5.2 ms.