Principal Investigator, Video Analytics

Professional Experience

I have led the design, development, and implementation of a range of successful data analytics systems, including radar signal and image processing, text analytics, and image and video analytics. In my text analytics LDRD, we developed a framework combining faceted navigation, random-walk graph analytics, and topic modeling that yielded a unique analyst-in-the-loop modeling and retrieval (AMR) capability. In my current video analytics LDRD, we are developing and experimenting with rich multimodal feature spaces that facilitate AMR for video data. We are also taking a lead role in developing a new kind of multimodal dataset ecosystem building on the YFCC-100M dataset, a collection of 99 million flickr images and 800,000 videos, which we partnered with Yahoo Labs and ICSI to create and publish.

Research Interests

My interest is in understanding how the brain perceives the outside world and using that as inspiration for new approaches to generating multimodal feature spaces along with new algorithms and interfaces for video analytics applications. I am currently focused on deep learning techniques, graph analytics, and dataset development.

M.S., Electrical Engineering, Stanford University, 2000

M.S., Nuclear Engineering, University of Michigan, 1987

B.S., Nuclear Engineering, University of Michigan, 1984

K. Ni, C. Carrano, D. Poland, B. Elizalde, G. Friedland, L. Gottlieb, and D. Borth, “The Yahoo-Livermore-ICSI (YLI) Multimedia Feature Set and YFCC100M Corpus,” http://www.yli-corpus.org, 2015.

B. Thomee, D. Shamma, G. Friedland, B. Elizalde, K. Ni, D. Poland, D. Borth, and L. Li, “The New Data and New Challenges in Multimedia Research,” arxiv:1503.01817v1 (also submitted to Communications of the ACM, currently undergoing revision for resubmission), 2015.

G. Freidland, C. Ngo, D. Shamma, D. Poland, B. Thomee, M. Larson, “Multimedia COMMONS – Community-Organized Multimodal Mining: Opportunities for Novel Solutions,” ACM Multimedia 2015: MMCommons Workshop, http://www.mmcommons.org, LLNL-PROP-667444, 2015.