PROJECT TITLE :
Depth Reconstruction From Sparse Samples: Representation, Algorithm, and Sampling
The rapid development of 3D technology and pc vision applications has motivated a thrust of methodologies for depth acquisition and estimation. But, existing hardware and software acquisition strategies have limited performance due to poor depth precision, low resolution, and high computational cost. In this paper, we gift a computationally economical technique to estimate dense depth maps from sparse measurements. There are three main contributions. First, we tend to provide empirical proof that depth maps can be encoded a lot of additional sparsely than natural images using common dictionaries, such as wavelets and contourlets. We conjointly show that a combined wavelet–contourlet dictionary achieves higher performance than using either dictionary alone. Second, we have a tendency to propose an alternating direction method of multipliers (ADMM) for depth map reconstruction. A multiscale warm begin procedure is proposed to hurry up the convergence. Third, we tend to propose a two-stage randomized sampling scheme to optimally opt for the sampling locations, so maximizing the reconstruction performance for a given sampling budget. Experimental results show that the proposed methodology produces high-quality dense depth estimates, and is sturdy to noisy measurements. Applications to real data in stereo matching are demonstrated.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here