Light Field Imaging: - A Comparative Study of Plenoptic Image Compression Techniques

Abstract

Richer visual information from our world can be captured using light field imaging that has emerged as a new trend. As opposed to traditional photography which captures a 2D projection of the light in the scene integrating the angular domain, light fields collect radiance from rays in all directions demultiplexing the angular information lost in conventional photography. Compression algorithms play a vital role in the efficient storage and delivery of a plenoptic image. Several methods to compress light field images have been proposed recently. However, in-depth evaluations of compression algorithms have rarely been reported. In this report, mainly three prediction techniques are compared. Firstly, micro image composing a plenoptic image are processed by an adaptive prediction tool, aiming at reducing data correlation before entropy coding takes place. In the second coding structure, LI minimization of the residuals is proposed. The disk-shaped pixel clusters corresponding to each micro lens in the light field image are efficiently predicted on the basis of neighboring disks. The captured images contain repetitive pattern that resulted from adjacent micro lenses. A full inter-prediction scheme in video image is introduced into intra-prediction for the compression of light field image is discussed in the third technique. All the three techniques are then compared based on their bit error rate(BER) and based on that values, the last coding technique where both inter and intra prediction is incorporated provides better results than the other two techniques.