Partial Coherence for Object Recognition and Depth Sensing
This work addresses the problem of optimizing illumination for computer vision systems, but it appears incremental as it applies existing computational and deep learning methods to a new parameter.
The paper investigates how varying degrees of illumination coherence affect computer vision tasks, finding that increased coherence improves image entropy and enhances object recognition and depth sensing performance.
We show a monotonic relationship between performances of various computer vision tasks versus degrees of coherence of illumination. We simulate partially coherent illumination using computational methods, propagate the lightwave to form images, and subsequently employ a deep neural network to perform object recognition and depth sensing tasks. In each controlled experiment, we discover that, increased coherent length leads to improved image entropy, as well as enhanced object recognition and depth sensing performance.