The research work in this project is focused on 3D video stream adaptation methods capable of maximizing 3D perceptual quality under different sets of known constraints. The joint interdependencies between rate, 3D perceptual distortion and decoding complexity will be investigated using subjective evaluation and validation procedures. Adaptation of 3D video streams is mainly targeted at rate, format and decoding complexity adaptation, taking into account that 3D video includes more parameters suitable for adaptation which provide an increased level of flexibility (e.g., the 3D signal is coded into two streams which might undergo through different channel errors and constraints, variable depth distortion in the spatial and temporal domains producing similar perceptual quality, asymmetric distortion in stereoscopic video, effect of error concealment methods in 3D perception, etc). The project research is also concerned in devising power-aware adaptation functions such that decoding power consumption is reduced at the expense of minimum quality cost. This is an open topic for research in 3D video since decoding is far more complex than monoscopic video and the need for low complexity decoding is higher because of limited battery life and lower computational resources in portable decoders.
|Start Date: 01-03-2011|
|End Date: 01-03-2014|
|Team: Pedro Antonio Amado Assunção, Sylvain Tony Marcelino|
|Groups: Multimedia Signal Processing – Lr|
|Local Coordinator: Pedro Antonio Amado Assunção|
|Links: Internal Page|