The recently introduced compressed sensing (CS) theory can, potentially, simplify the acquisition process of resource-limited devices by taking advantage of signal sparsity. This paper proposes a perceptual-based compressed video sensing (CVS) strategy that benefits from the human visual perception properties. Two frameworks are proposed, namely, Intraperc-CVS and Inter-perc-CVS. In both frameworks, an efficient perceptualbased weighting strategy is applied for acquisition and recovery. In the Intraperc-CVS scheme, video frames are acquired and recovered separately, while in the Inter-perc-CVS scheme, the frames are recovered jointly to further exploit inter-frame correlation. The proposed perceptual-based frameworks show remarkable performance improvement over the standard CVS.

A perceptual-based compressed sensing (CS), which focuses the measurements and the recovery on the visually important low-frequency coefficients, is applied for multi-view image signals. High correlation among different views is exploited to generate signal prediction using disparity estimation and compensation techniques. A residual-based recovery is utilised as a joint recovery for the nonreference images to enhance the reconstruction performance. The proposed framework shows remarkable performance improvement over the conventional CS with joint recovery as well as the perceptualbased CS with independent recovery.

Exploiting perceptual-based weighting can improve the reconstruction quality for compressed video sensing (CVS). Nevertheless, practical implementation of the compressed sensing (CS) requires quantizing the measurements. We propose a simplified sampling rate model for the perceptual CVS to achieve compromise between the number of measurements and the quantization bit-depth, which are the main contributing factors in the CS rate-distortion (RD) performance. The proposed model can achieve near optimal RD-performance obtained through exhaustive simulations. In addition, simulation results show that the quantized perceptual CVS achieves remarkable RD-performance gain, with lower sampling rate, compared to applying the quantization model on the standard CS.

This paper proposes an approach of compressed sensing (CS) of video in which distributed video coding DVC and CS are integrated as in [1], and the sensing matrix is modulated in suit of [2] but with proposed fixed weighting strategy to certain DCT coefficients in an effort to improve the visual quality of reconstruction.