Compressive video sensing using non-linear mapping

Compressive sensing provides a formalized mathematical framework to acquire and reconstruct sparse signals using sub-Nyquist sampling rate, and has great potential in the application of image and video acquisition and compression. In this paper, by incorporating improved OMP algorithm via non-linear mapping, our proposed compressive video sensing framework has the advantages of lower complexity than that of other convex optimization based framework, and improved reconstruction performance compared with traditional OMP algorithm. Experimental results have demonstrated the effectiveness of our framework.