Parallel@Illinois Special Seminar Series

Multi-Frame Video Enhancement: A Better Video for Everyone

Wednesday, April 1st at 4:15 pm CT, B02 CSL

Nikola Bozinovic, Vice President of Engineering, MotionDSP

ABSTRACT: MotionDSP makes software that significantly improves video from a wide range of sources -- from mobile phones to surveillance cameras. Its patented, multi-frame video enhancement technology is available in Ikena, a video forensic solution, and in vReveal, a Windows application for consumers.

The open-loop nature of MotionDSP's multi-frame video reconstruction framework requires a departure from widely used quantized block-based motion estimation concept. Unlike video encoding, super-resolution can not rely on a prediction error to compensate for motion estimation inaccuracies. The need to use floating-point accurate motion fields, coupled with a multi-frame nature of the algorithm, makes GPGPU a natural choice when addressing this massive computational problem.

This talk will provide an overview of MotionDSP's technology and products, with an emphasis on robust and highly accurate motion modeling and estimation. It will also provide details on the design and performance challenges of MotionDSP's video pipeline, the first commercial video processing solution fully ported to GPU architecture. Along with the existing forensic and personal video enhancement applications, new applications in areas of video conferencing and Internet video will be discussed.

BIO: Nikola Bozinovic is a VP of Engineering at MotionDSP. Under his technical lead, MotionDSP partnered with In-Q-Tel and NVIDIA to bring the next-generation video enhancement technology to both intelligence community and general consumers. Prior to starting MotionDSP's R&D in 2006, he held engineering positions at Microsoft Research Asia and Veodia Inc. He was awarded EURASIP best paper award (co-authored with Janusz Konrad) for his work on frequency-domain motion analysis. His other research interests include video reconstruction and restoration, motion modeling, and video analytics. He holds a Ph.D. in Electrical and Computer Engineering from Boston University, and a B.S.E.E. from University of Nis, Serbia.

For more information about MotionDSP, visit their website.



Emerging Application Domains

  • Computer vision
  • Graphics
  • Gaming/Virtual worlds
  • Robotics/Autonomous vehicles
  • Machine learning
  • Speech processing
  • Human-computer Interfaces
  • Computational Science
  • Biomedical computing
  • Imaging/Video

Related links

Computer Science

Coordinated Science Laboratory

Electrical and Computer Engineering

Parallel@Illinois

Parallel@Illinois