Optical Neural Networks
This is an incremental improvement for computer vision applications, enhancing accuracy in image classification tasks.
The paper tackles image classification by introducing an optical neural network (ONN) framework that uses multi-level zoom based on a focal point, inspired by the human eye, to achieve high accuracy improvements without altering underlying CNN architectures.
We develop a novel optical neural network (ONN) framework which introduces a degree of scalar invariance to image classification estima- tion. Taking a hint from the human eye, which has higher resolution near the center of the retina, images are broken out into multiple levels of varying zoom based on a focal point. Each level is passed through an identical convolutional neural network (CNN) in a Siamese fashion, and the results are recombined to produce a high accuracy estimate of the object class. ONNs act as a wrapper around existing CNNs, and can thus be applied to many existing algorithms to produce notable accuracy improvements without having to change the underlying architecture.