Authors: Mithilesh Kumar Jha; Sumantra Dutta Roy; Brejesh Lall
Addresses: Department of Electrical Engineering, Indian Institute of Technology, Delhi, India ' Department of Electrical Engineering, Indian Institute of Technology, Delhi, India ' Department of Electrical Engineering, Indian Institute of Technology, Delhi, India
Abstract: The new generation multimedia applications such as ultra-high definition TV, 3D video and cloud storage keeps the compression research relevant even post to the advent of the state-of-art standard HEVC (High Efficiency Video Coding) codec. The continued studies on human visual system suggest that a perceptually acceptable reconstruction with reduced bandwidth requirements may be more acceptable and commercially viable solution rather than a bit accurate reconstruction for a variety of video and graphics applications. One way to achieve compression beyond that achieved by the advanced codec like HEVC is to sample the input data at a lower rate than the Nyquist rate, exploit the sparsity and perceptual redundancy in the content using advanced signal processing and computer vision tools and represents it using a spatio-temporal model. This paper contains an exhaustive review of the new context of video compression methods along with their limitations and open issues.
Keywords: compressive sensing; CS; 3D mosaic; directional empirical mode decomposition; DEMD; high dynamic range; HDR; high frequency rate; HFR; parametric video compression; PVC; statistically matched wavelet; SMWT; texture video; ultra-high definition television; UHD TV; wide colour gamut; WCG.
International Journal of Signal and Imaging Systems Engineering, 2017 Vol.10 No.4, pp.204 - 222
Received: 08 Jun 2016
Accepted: 09 May 2017
Published online: 21 Aug 2017 *