METHOD AND SYSTEM OF CONVOLUTION IN NEURAL NETWORK WITH VARIABLE DILATION RATE

• Main Authors: YADAV, Dinesh Kumar, DAS, Saptarsi, LEE, Sehwan, JANG, Junwoo, DESHWAL, Ankur
• Format: Patent
• Language: English; French; German
• Published: 09.08.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The present invention relates to a method of convolution in a neural network with variable dilation rate, the method includes receiving size of a first kernel, and a dilation rate. Further, at least one of size of one or more disintegrated kernels is determined based on the size of the first kernel, a baseline architecture of a memory and the dilation rate. Furthermore, an address of one or more blocks of an input image is determined based on the dilation rate, and one or more parameters associated with a size of the input image and the memory. Thereafter, the one or more blocks of the input image and the one or more disintegrated kernels are fetched from the memory. Finally, the method includes obtaining an output image based on convolution of each of the one or more disintegrated kernels and the one or more blocks of the input image.