PhD Thesis Defense
Augmenting Cartographic Resources and Assessing Roadway State for Vehicle Navigation
Event Location: NSH 1507Abstract: Maps are important for both human and robot navigation. Given a route, driving assistance systems consult maps to guide human drivers to their destinations. Similarly, topological maps of a road network provide a robotic vehicle with information about where it can drive and what driving behaviors it should use. Maps simplify [...]
Lumenhancement: Exploiting Appearance for Planetary Modeling
Event Location: NSH 1109Abstract: Planetary environments are among the most hazardous, remote and uncharted in the solar system. They are also critical to the search for life, human exploration, resource extraction, infrastructure and science. These applications represent the prime unexploited opportunity for automated modeling, but robots are under-utilized for this purpose. There is urgent need [...]
Attention-guided Algorithms to Retarget and Augment Animations, Stills, and Videos
Event Location: GHC 4301Abstract: Still pictures, animations and videos are used by artists to tell stories visually. Computer graphics algorithms create visual stories too, either automatically, or by assisting artists. Why is it so hard to build algorithms that can manipulate the content created by a visual artist? A primary reason is that artists introspect [...]
Fast and Graceful Balancing Mobile Robots
Event Location: NSH 1305Abstract: Personal mobile robots will soon be operating and closely interacting with us in human environments. Balancing mobile robots can be effective personal robots as they can be tall enough for eye-level interaction and narrow enough to navigate cluttered environments, and they also have the dynamic capabilities to move with speed and [...]
Distributed Planning Under Uncertainty for Large Teams
Event Location: NSH 1305Abstract: In many domains, teams of hundreds of agents must cooperatively plan to perform tasks in a complex, uncertain environment. This requires that each agent take into account teammates' states, observations, and actions when making decisions about its own actions. Naively, finding a good policy involves searching this large joint space, but [...]
Vision-Based Control of a Handheld Micromanipulator for Robot-Assisted Retinal Surgery
Event Location: NSH 1507Abstract: Surgeons increasingly need to perform complex operations on extremely small anatomy. Many existing and promising new surgeries are effective, but difficult or impossible to perform because humans lack the extraordinary control required at sub-millimeter scales. Using micromanipulators, surgeons gain higher positioning accuracy and additional dexterity as the instrument removes tremor and [...]
Context and Subcategories for Sliding Window Object Recognition
Event Location: NSH 3305Abstract: Object recognition is one of the fundamental challenges in computer vision, where the goal is to identify and localize the extent of object instances within an image. The current de facto standard for building high-performance object category detectors is the sliding window approach. This approach involves scanning an image with a [...]
Lifelong Robotic Object Recognition
Event Location: NSH 3305Abstract: In this thesis, we study the topic of Lifelong Robotic Object Perception. We propose, as a long-term goal, a framework to recognize known objects and to discover unknown objects in the environment as the robot operates, for as long as the robot operates. We build the foundations for Lifelong Robotic Object [...]
Model Recommendation for Action Recognition and Other Applications
Event Location: GHC 4405Abstract: The typical approach to learning based vision has been that for each individual application, classifiers or detectors are learned anew from annotated training data for each specific task. However, the classifiers trained in this way tend to be brittle and highly specialized to the datasets from which they are derived, making [...]
Object Instance Discovery from Scenes of Daily Living
Event Location: NSH 1507Abstract: This thesis tackles the problem of automatically discovering objects from a collection of images from the Activities of Daily Living (ADL) environment. We contribute, 1) a framework for discovering object instances under severe clutter, occlusion, changes of view point, heterogeneity of object appearance and imperfect segmentation; 2) a data-driven approach for [...]