Content adaptive video processing algorithms for digital tv

  • Authors:
  • Mainak Biswas;Truong Q. Nguyen

  • Affiliations:
  • University of California, San Diego;University of California, San Diego

  • Venue:
  • Content adaptive video processing algorithms for digital tv
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

The advent of high definition digital television systems has changed the way we view content. The increased resolution of the broadcast video coupled with better displays in the form of Liquid Crystal Displays (LCD) and Plasma Display Panel (PDP) have made the viewing experience pleasant. Penalties that must be paid for these new technology are in terms of increased complexity of the video signal processing, the visibility of the hitherto imperceptible artifacts and cost. The processing of video signals for display on these new panels is a big challenge. Traditional TV manufacturers have never realized the need to de-interlace, since a Cathode Ray Tube (CRT) could display the interlaced signal directly. Similarly, image scaling/resolution changes are accommodated on a CRT by adjusting the scan size. With the new displays, which have fixed rasters addressed sequentially, the TV manufacturers have suddenly found the need to incorporate de-interlacing and scaling technologies. Due to large display size, artifacts such as MPEG compression artifacts which were not noticeable in smaller display have become unbearable to watch. The LCDs and PDPs have different display characteristics, that need separate processing such as Response Time Compensation (RTC) and motion de-blurring for better motion portrayal. In this dissertation, some of the solutions that mitigate these problems are proposed. "True" motion information are extracted from the video sequences by mixed model motion estimator and the information are used in the later stages for de-interlacing, frame rate conversion, video in-painting and de-convolution to reduce the hold type motion blur in LCD. ν-SVM based noise removal algorithm is proposed that reduces blocking, ringing and mosquito artifact from compressed video. The use of these technologies goes a long way to realize the full potential of the high definition TV systems.