Seminar Series Archive

Computer vision (CV) is the area of study dedicated to helping computers see and understand the meaning in images and videos captured using cameras. Until recently, this field was dominated by hand crafted specialized models and algorithms using signal processing and statistical methods. Within the last decade, advances in deep learning has enabled a major cross-fertilization between computer vision and machine learning fields with state-of-the art results on challenging CV tasks. In this talk, I will provide insights on these results for three tasks critical for many real world applications: object recognition, object detection and depth estimation. I will conclude by highlighting the limitations and challenges that still remain to enable computers to truly see and understand in real world scenarios.

Narayan Srinivasa is an expert in machine learning and neuromorphic computing and its applications to solve real world problems. He will be joining Intel Labs as Director of Machine Intelligence research in July 2019, Until recently, he was the CTO of Eta Compute, developing energy efficient audio applications for edge devices. Prior to Eta Compute, Dr. Srinivasa worked at Intel Labs as Chief scientist and Senior Principal Engineer and led the development of the 14 nm Loihi neuromorphic chip with on-chip learning. Before Intel, he was the Director for the Center for Neural and Emergent Systems at HRL and served in various capacities including the principal investigator for DARPA programs SyNAPSE, Physical Intelligence, UPSIDE, and others. Dr. Srinivasa has 63 issued patents and published over 90 articles in peer-reviewed journals and conferences. Reports about his work have appeared in The Economist, MIT Technology Review, Wired Magazine, IEEE Spectrum and Forbes among others. Dr. Srinivasa has a Ph. D. from the University of Florida and was a Beckman Post-Doctoral Fellow at the University of Illinois at Urbana-Champaign.