Mapping the World with Street View

Luc Vincent, Google
Wednesday, November 5

Abstract: Street View started as a Google 20% project and quickly became one of the company's "moonshots". The long term dream is to capture street level imagery everywhere possible, organize it, and make it universally accessible and useful. Towards this ambitious goal, the team has already published imagery from over 60 countries, and is continuing to expand. In addition, this imagery is quickly becoming indispensable to the process of making high-quality maps at global scale. In this talk, I will explain how modern computer vision techniques based on deep learning are key to this process. They are themselves made possible by enormous datasets of training data, created using crowdsourcing techniques.

Biography: Luc Vincent is currently an engineering director in charge of imagery for Google's Geo products - including Street View. Luc earned his B.S. from Ecole Polytechnique, M.S. in Computer Science from University of Paris XI, and PhD in Mathematical Morphology from Ecole des Mines de Paris in 1990. He has over 60 publications in the area of computer vision, image analysis, and document understanding. He has served as an Associate Editor for the IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), and for the Journal of Electronic Imaging. He has also chaired SPIE's conferences on Document Recognition, the International Symposium on Mathematical Morphology (ISMM), and been in the program committee of numerous conferences and workshops. He joined Google a decade ago to work on the Google Books project. While he was ramping up Google's OCR efforts, he got involved in an early stage project whose goal was to capture a large amount of street level imagery and make it universally accessible and useful. Under his leadership, this project became Google Street View and launched officially in May 2007. Before Google, he was Chief Scientist, and then VP of Document Imaging at LizardTech, a developer of advanced image compression software. Prior to this, he led an R&D team at the prestigious Xerox Palo Alto Research Center (PARC). He was also Director of Software Development at Scansoft (now Nuance) and held various technical management and individual contributor positions at Xerox Corporation.

Interactive Crowd Simulation for Spatial Analysis of Indoor and Outdoor Environments

Dinesh Manocha, Department of Computer Science, University of North Carolina at Chapel Hill
Thursday, November 6

Abstract: A key issue in design of buildings and cities is to model how people behave and interact with spaces. This includes the ability to model, simulate and predict pedestrian movements, with all the pressures, delay actions and reactions that large groups of humans can generate. The ultimate goal is to design smart spaces that are optimized to enhance pedestrian utilization, efficiency, and human experiences.

In this talk, we give an overview of our work on simulating large crowd movements and behaviors in indoor and outdoor environments. These include velocity-based optimization algorithms that are combined with global wayfinding techniques. We use techniques from biomechanics, robotics, and psychology to accurately simulate human movement. The resulting algorithms map well to current multi-core and many-core processors and have been used to predict the movement of thousands of pedestrians in architectural and urban settings.

Biography: Dinesh Manocha is currently the Phi Delta Theta/Mason Distinguished Professor of Computer Science at the University of North Carolina at Chapel Hill. He has co-authored than 380 papers in the leading conferences and journals on computer graphics, robotics, and scientific computing. He has also served program chair for many conferences and editorial boards of many leading journals. Some of the software systems related to collision detection, GPU-based algorithms and geometric computing developed by his group have been downloaded by more than 150,000 users and are widely used in the industry. Manocha has received awards including Alfred P. Sloan Fellowship, NSF Career Award, Office of Naval Research Young Investigator Award, and 14 best paper awards at the leading conferences. He is a Fellow of ACM, AAAS, and IEEE, and received Distinguished Alumni Award from Indian Institute of Technology, Delhi.