Compression of Plenoptic Point Cloud Attributes Using 6-D Point Clouds and 6-D Transforms

In this paper, we introduce a novel 6-D representation of plenoptic point clouds, enabling joint, non-separable transform coding signals defined along both spatial and angular (viewpoint) dimensions. This representation, which is built in global coordinate system, can be used multi-camera studio capture video fly-by scenarios, with various viewpoint (camera) arrangements densities. We show that the Region-Adaptive Hierarchical Transform (RAHT) Graph Fourier (GFT) extended to proposed enable coding. Our method applicable data either dense or sparse sets viewpoints, complete xmlns:xlink="http://www.w3.org/1999/xlink">incomplete data, while state-of-the-art RAHT-KLT method, separable dimensions, only data. The “complete” refers has, for each point, one colour every (ignoring any occlusions), “incomplete” has colours xmlns:xlink="http://www.w3.org/1999/xlink">visible surface points at viewpoint. demonstrate RAHT GFT compression methods are able outperform on 3-D objects levels specularity, captured different camera degrees sparsity.