Jonathan Mooser PhD Student · Computer Science University of Southern California CGIT Lab mooser (at) graphics (dot) usc (dot) edu 323.363.9782 Resume

		Computer Vision and Object Recognition  Reliably identifying objects, places, and people in images is one of the most challenging problems in computer science. The difficulty stems largely from the fact that a single object may appear very different when viewed from different angles, under different lighting conditions, or when partially covered by another object. We explore recognition using visual features based on scale invariant feature transforms (SIFT) and Histograms of Oriented Gradients (HOG).
		Segmentation and Tracking Besides object recognition, another challenging problem is segmentation, which means accurately separating object from background. Our research focuses on the popular set of techniques based on max-flow/min-cut algorithms. Once an object has been segmented, the process of optical flow can be used to track it as either the camera or the object itself moves.
		Mobile Agmented Reality  Agmented Reality attempts to incorporate virtual objects into an image of the real world. While a number of methods have been suggested to track the camera's movements through the world, we focus on visual tracking. Either artificial markers or preexisting objects can be used to estimate the camera position and orientation at any time.   In an industrial context, a mobile device equipped with a camera can be used to place repair or maintenance information on top of an image of a piece of equipment.
		Facial Animation (Previous work) Improving blendshape animation by providing tools to alliviate the problem of "blendshape interference."

		CiSoft - A joint research effort of the University of Southern California and Chevron. Our goal is to develop interactive “smart oilfield” technologies. The CGIT lab, in conjunction with CiSoft, developes mobile software that allows field technicians to retreive equipment data and repair histories in real-time.
		Chojo - Chôjô combines handheld computing, GPS tracking, and wireless connectivity to create a multi-user game in which players collaborate to populate the USC campus with virtual plants.

J. Mooser, Q. Wang, S. You, and U. Neumann, “Fast Simultaneous Tracking and Recognition Using Incremental Keypoint Matching”, in Proceedings of the Fourth International Symposium on 3D Data Processing, Visualization and Transmission (3DPVT) , June 2008.

J. Mooser, S. You, and U. Neumann, “Large Document, Small Screen: A Camera Driven Scroll and Zoom Control for Mobile Devices”, in Proceedings of ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games (I3D), pp. 27-34, February 2008.

J. Mooser, S. You, and U. Neumann, “Real-Time Object Tracking for Augmented Reality Combining Graph Cuts and Optical Flow”, in International Symposium on Mixed and Augmented Reality, pp. 145-152, November 2007.

Q. Wang, J. Mooser, S. You, and U. Neumann, “Augmented Exhibitions Using Natural Features”, in ISMAR 2007 Workshop on Mixed Reality Entertainment and Art, pp. 27-32, November 2007.

J. Mooser, L. Wang, S. You, and U. Neumann, “An Augmented Reality Interface for Mobile Information Retrieval”, in International Conference on Multimedia and Expo, pp.2226-2229, July 2007.

J. Mooser, S. You, and U. Neumann, “Tricodes: A Barcode-Like Fiducial Design for Augmented Reality Media”, in International Conference on Multimedia and Expo, pp.1301-1304, July 2006.

J. P. Lewis, J. Mooser, Z. Deng, and U. Neumann, “Reducing Blendshape Interference by Selected Motion Attenuation.” in Proceedings of ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games (I3D), pp. 25-29, April 2005.