About This Webinar
Depth or 3D imaging refers to image sensors that measure the distance between the camera and the target object on a pixel-by-pixel basis. While commercial depth cameras come in many shapes and sizes while relying on a host of modalities to make these measurements, there is a growing body of literature focused on using AI to estimate depth from tradition area scan cameras, either monochrome or color.
The applications for these AI models are profound as area scan cameras are far less expensive than 3D; however, they move the expense of sensing to the expense or processing. For instance, self-driving cars are evolving to remove their LIDAR as well as sonar sensors. So in this talk, Lau reviews a selection of AI models that have been proposed over the years and discusses the means by which they collect and label data for training as well as take a glimpse at the quality of 3D that they produce.
*** This presentation premiered during the
2024 Vision Spectra Conference. For more information on Photonics Media conferences and summits, visit
events.photonics.com
About the presenter
Daniel L. Lau, Ph.D., received his B.Sc. degree with highest distinction in electrical engineering from Purdue University, West Lafayette, IN, in 1995 and the doctorate degree from the University of Delaware, Newark, in 1999. Today, he is the Databeam professor and director of graduate studies at the University of Kentucky, Lexington. His research interests are in signals and systems with particular interests in 3D imaging for machine vision applications as well as multi and hyper spectral imaging through coded aperture spectral snapshot imaging and compressive sensing. Lau's published works include articles in the Proceedings of the IEEE, IEEE Transactions on Pattern Analysis and Machine Intelligence, IEEE Signal Processing Magazine, as well as Elsevier Signal Processing journal.
His experience in teaching includes classes in analog and digital communications as well as electronic devices and circuits. Outside of academia, Dr. Lau was a DSP Engineer at Aware, Inc., and an Image and Signal Processing Engineer at Lawrence Livermore National Laboratory. Lau has consulted with industry sponsors including Intel, Valvoline, and Agere Systems. He’s a founding member of Seikowave Inc, developing 3D scanners for nondestructive testing of oil and gas pipelines. Lau is also a member of the Association for Advanced Automation (A3) technical committee for machine vision as well as their newly formed committee for workforce education and training.