Abstract: Examples disclosed herein include a video-data encoder. The video data encoder may encode a 4×4 data block into a bit stream according to a context adaptive variable length coding. The 4×4 data block may be representative of video data. The video-data encoder may, while encoding the 4×4 data block, ignore at least some coefficients of the 4×4 data block. In some examples, the video-data encoder may ignore the at least some coefficients of the 4×4 data block by setting the at least some coefficients of the 4×4 data block to zero prior to encoding the 4×4 data block. Related devices, systems and methods are also disclosed.

Abstract: A method and apparatus for performing a convolution of a N×N matrix. A weights matrix for a N×N Convolutional Neural Network (CNN) is received and is divided into 3×3 weights matrixes. Lines of image values are read and are stored in a buffer as sets of image values. A 3×3 convolution is performed to generate a 3×3 convolution value. All 3×3 convolution values associated with a particular N×N convolution and a particular set of image values are summed. The 3×3 convolutions and the summing are repeated until all columns in the set of image values have been processed; and the reading, the storing, the performing 3×3 convolutions, the summing and the repeating performing 3×3 convolutions are repeated until all lines of image values have been processed. The sums associated with a particular N×N convolution are added together to generate an N×N convolution value for each of the N×N convolutions.