Classification of optics-free images with deep neural networks

11/10/2020

∙

by   Soren Nelson, et al.

∙

17

∙

share

The thinnest possible camera is achieved by removing all optics, leaving only the image sensor. We train deep neural networks to perform multi-class detection and binary classification (with accuracy of 92 images without the need for anthropocentric image reconstructions. Inferencing from optics-free images has the potential for enhanced privacy and power efficiency.