Seminar
Object Detection and Tracking on Low Resolution Aerial Images
Abstract: Object tracking from an aerial platform poses a number of unique challenges including the small number of pixels representing the objects, large camera motion, and low temporal resolution. Because of these unique reasons, low resolution aerial image analysis needs to be tackled differently than the traditional image analysis both in terms of the sensors, [...]
From Automation to Autonomy and the Ubiquity of Moral Decision Making
Abstract: I argue that there is an important sense in which all decisions are moral decisions and I explore some implications of this insight (and its denial) for the design and human impacts of increasingly complex automated systems and emerging autonomous systems. This insight is obscured when we think about automated systems by the social [...]
Personalized model of kinematic and dynamic of physical activities
Teruko Yata Memorial Lecture in Robotics Abstract: By now it has become a cliché the statement that the population in industrial world is aging and hence the problem of physical agility is a serious health problem. Moreover this issue is aggravated even with younger population due to our sedative life style. It also is an [...]
Splitting the Equity Pie & Other Aspects of Equity Compensation
Abstract: Equity compensation can be a very mysterious concept for first-time entrepreneurs. Learn how to make critical decisions from the division of the Founders' Pie through stock option grants & vesting schedules. Biography: Frank Demmler is Innovation Works’ Vice President of The Entrepreneurial Services Team, the internal group that assists early stage companies’ efforts [...]
Data-Driven Learning Towards Perceptual Organization
Abstract: Computer vision has advanced rapidly with deep learning, achieving above human performance on some classification benchmarks. At the core of the state-of-the-art approaches for image classification, object detection, and semantic/instance segmentation is sliding window classification, engineered for computational efficiency. Such piecemeal analysis of visual perception often has trouble getting details right and fails miserably [...]
Learning to Drive
Abstract: Why is our understanding of sensorimotor control behind our understanding of perception? I will talk about structural differences between perception and control, and how these differences can be mitigated to help advance sensorimotor control systems. Judicious use of simulation can play an important role and I will describe some simulation tools that we have [...]
Composable Benchmarks for Safe Motion Planning on Roads
Abstract Numerical experiments for motion planning of road vehicles require numerous components: vehicle dynamics, a road network, static obstacles, dynamic obstacles and their movement over time, goal regions, a cost function, etc. Providing a description of the numerical experiment precise enough to reproduce it might require several pages of information. Thus, only key aspects are [...]
Deep Representation Learning with Induced Structural Priors
Abstract: With the support of big-data and big-compute, deep learning has reshaped the landscape of research and applications in artificial intelligence. Whilst traditional hand-guided feature engineering in many cases is simplified, the deep network architectures become increasingly more complex. A central question is if we can distill the minimal set of structural priors that can [...]