Home > Ryad Benosman - What is Neuromorphic Event-based Computer Vision? Sensors, Theory and Applications

Ryad Benosman - What is Neuromorphic Event-based Computer Vision? Sensors, Theory and Applications

Start:

9/12/2019 at 3:30PM

End:

9/12/2019 at 4:45PM

Location:

131 DeBartolo

Host:

College of Engineering close button
headerbottom

Siddharth Joshi

Siddharth Joshi

VIEW FULL PROFILE Email: sjoshi2@nd.edu
Phone: 574-631-8380
Website: http://www.siddharth-joshi.com
Office: 326B Cushing Hall
Curriculum Vitae

Affiliations

Wireless Institute Assistant Professor
Dr. Joshi's research develops extremely power efficient processors and circuits that can operate under severe resource constraints  such as sensors and processors used in the “Internet of Things” IOT. His group  explores hardware and software techniques  to enable adaptive and learning algorithms ...
Click for more information about Siddharth
574-631-8380
Add to calendar:
iCal vCal

In this presentation, I will introduce neuromorphic, event-based approaches for image sensing and processing. State-of-the-art image sensors suffer from severe limitations imposed by their very principle of operation. These sensors acquire the visual information as a series of “snapshots” recorded at discrete point in time, hence time-quantized at a predetermined frame rate, resulting in limited temporal resolution, low dynamic range and a high degree of redundancy in the acquired data.  Nature suggests a different approach: Biological vision systems are driven and controlled by events happening within the scene in view, and not – like conventional image sensors – by artificially created timing and control signals that have no relation to the source of the visual information.

Translating the frameless paradigm of biological vision to artificial imaging systems implies that control over the acquisition of visual information is no longer imposed externally on an array of pixels but rather the decision making is transferred to each individual pixel, which handles its own information individually. We will introduce the fundamentals underlying such bio-inspired, event-based image sensing and processing approaches, and explore their strengths and weaknesses.  I will show that bio-inspired vision systems have the potential to wipe out conventional, frame-based vision acquisition and processing systems and to establish new benchmarks in terms of  data compression, dynamic range, temporal resolution and power efficiency in applications such as 3D vision, object tracking, motor control, visual feedback loops, and machine learning in real-time at several hundreds kHz.

Seminar Speaker:

Ryad Benosman

Ryad Benosman

Ryad B. Benosman is a full Professor at both the University of Pittsburgh Medical Center, Carnegie Mellon University and Sorbonne Universitas where he does research at the intersection of robotics, computer vision and neuroscience. Specifically, he investigates the use of standard and neuromorphic cameras to enable autonomous, agile robotics, brain-machine interfaces focusing on retina prosthetics, optogenetics stimulation and recently visual cortex stimulation.

Ryad did his PhD in robotics and computer vision at University of Pierre and Marie Curie after studying pure and applied mathematics. He is a pioneer and a leading researcher in the field of event based neuromorphic computer vision. His lab developed the ATIS neuromorphic camera. He has authored more than 200 papers, 60 of which are considered to provide the foundations of neuromorphic computer vision. He also founded several companies such as Prophesse (formerly Chronocam) the leading company in event-based vision, Pixium Vision (retina prosthetics), Chronolife (eHealth) and more recently Brainiac (neural processor computer).