Emma Alexander

ealexander@northwestern.edu

Google Scholar

CV

I'm interested in low-level, physics-based, bio-inspired artificial vision.
For a high-level overview  of my research approach see my ACM XRDS article.

Computational imaging combines the design of optical systems with post-processing algorithms to reveal information that is invisible to standard cameras. In my lab, this information includes scene depth and high-dimensional spectral colors, as well as medical and scientific quantities related to human health, brain function, and cosmological structure. We use explicit physical models of image formation and draw inspiration from nature: all biological vision is computational imaging, because eyes and brains cooperate. For half a billion years, evolutionary processes have isolated promising regions of the design space across a diverse catalog of high-performing and tightly-constrained visual systems. These systems continue to outperform the technological state of the art in their robustness, compactness, speed, and power consumption. The importance of these advantages is growing rapidly with emerging demands for sustainable and edge-capable AI. By combining the engineering toolkit of computational imaging with the reverse engineering approach of vision science, my research connects algorithm and system design with basic science in a mutually-informative loop for robust and efficient computational imaging.