RT Conference Proceedings
T1 Filter optimization and complexity reduction for video coding using graph-based transforms
A1 Martínez Enríquez, Eduardo
A1 Díaz de María, Fernando
A1 Cid Sueiro, Jesús
A1 Ortega Gómez, Román Antonio
AB The basis functions of lifting transform on graphs are completelydetermined by finding a bipartition of the graph and defining theprediction and update filters to be used. In this work we consider thedesign of prediction filters that minimize the quadratic prediction errorand therefore the energy of the detail coefficients, which will giverise to higher energy compaction. Then, to determine the graph bipartition,we propose a distributed maximum-cut algorithm that significantlyreduces the computational cost with respect to the centralizedversion used in our previous work. The proposed techniquesshow improvements in coding performance and computational costas compared to our previous work.
PB IEEE
SN 978-1-4799-2341-0
YR 2013
FD 2013
LK https://hdl.handle.net/10016/18793
UL https://hdl.handle.net/10016/18793
LA eng
NO This work was supported in part by NSF under grant CCF-1018977
DS e-Archivo
RD 1 jul. 2024