Please briefly describe your background in imaging and vision.
Back in my grad school days, we used to mount sensors on mobile robots in a way that mimicked the eyes or antennae of insects. That was my first experience in computer vision -- enabling robots to extract information from images in order to navigate their environment. Since then, I've worked on human face analysis, visual tracking, and, most recently, 3D stereo depth sensing for automotive safety and gesture interfaces.
Your presentation will talk about computer vision - can you define computer vision and how the sensor requirements differ from applications with a display output?
In computer vision we extract targeted information from a scene and make it available for decision-making by machines. Unlike imaging and video, this information is not intended for human visual consumption. Typically, sensors for computer vision applications are selected to solve a specific problem, such as detecting and recognizing street signs.
Computer vision technology is in the early stages of development. What are the key focus areas for research at present?
This is actually the subject of my talk at IS Americas. Computer vision is enabling us to transition from "sensing" our environment to "understanding" it. Currently, there is a lot of research activity in object detection, such as pedestrian detection and automatic text reading for the visually impaired. Another area receiving a lot of attention is face detection, particularly for the automated curation of photo albums. Place recognition is one of the hottest trends in the augmented reality field. And there is also a lot of human analysis happening, including gesture recognition and the automated analysis of video security footage.
What are the current and potential future applications for this technology from a consumer and industrial standpoint?
Consumers have already adopted automotive safety features that rely heavily on computer vision; self-driving cars are in the works. As for manufacturing, machine inspection systems have proven invaluable, and vision will play a key role in next-gen assembly robots that learn by seeing.
Finally, we are delighted to have you participating as a speaker. What are you hoping to gain from your attendance at ISA?
I look forward to learning the latest in image sensors from experts in the field and hope to share a bit more about computer vision with you all.