A New Reconstruction Filter for Undersampled Light Fields

Abstract

This paper builds on previous research in the light field area of image-based rendering. We present a new reconstruction filter that significantly reduces the "ghosting" artifacts seen in undersampled light fields, while preserving important high-fidelity features such as sharp object boundaries and view-dependent reflectance. By improving the rendering quality achievable from undersampled light fields, our method allows acceptable images to be generated from smaller image sets. We present both frequency and spatial domain justifications for our techniques. We also present a practical framework for implementing the reconstruction filter in multiple rendering passes. Keywords: Image-based rendering, light field, lumigraph, sampling, reconstruction, aliasing, multipass rendering.