Uncanny Vision is focused on delivering accelerated computer vision solutions for the embedded world. Our first product UUncannyCV is an image-processing/computer-vsion library optimized for Cortex A series ARM processors. UncannyCV uses Neon instructions, algorithm optimizations and data prefetching to achieve performance boosts varying from 2x to 20x.
UncannyCV APIs can be loosely classified into 3 Low level APIs - convolution, morphological, rescaling, rotate/flip, pyramid, histogram, colour conversion etc Mid level APIs - Canny, Hough, K-means clustering, Lens distortion correction, Harris Corner, FAST, connected components etc High level APIs - Optical flow, ORB, Homography estimation(Ransac method) etc New algorithms are continuously getting added to UncannyCV. Pedestrian detection using HOG(Histogram of Oriented Graidents) is currently under development. UncannyCV supports SDKs for Linux, Android and IOS. A version for RTOSes is also getting ready.