End-to-End Framework for Efficient Deep Learning Using Metasurfaces Optics
This work addresses the challenge of deploying deep learning in resource-constrained environments for applications requiring efficient CNN processing.
This paper proposes an end-to-end framework that uses metasurface optics to optically compute Convolutional Neural Networks (CNNs) in free-space. This approach allows for direct processing of RGB data from natural scenes, achieving up to an order of magnitude energy saving with minimal impact on network accuracy.
Deep learning using Convolutional Neural Networks (CNNs) has been shown to significantly out-performed many conventional vision algorithms. Despite efforts to increase the CNN efficiency both algorithmically and with specialized hardware, deep learning remains difficult to deploy in resource-constrained environments. In this paper, we propose an end-to-end framework to explore optically compute the CNNs in free-space, much like a computational camera. Compared to existing free-space optics-based approaches which are limited to processing single-channel (i.e., grayscale) inputs, we propose the first general approach, based on nanoscale meta-surface optics, that can process RGB data directly from the natural scenes. Our system achieves up to an order of magnitude energy saving, simplifies the sensor design, all the while sacrificing little network accuracy.