Progressive lossless 3D mesh encoder with octree-based space partitioning

A progressive lossless 3D mesh encoder using the octree-based space partitioning of vertex data is proposed. Given a 3D mesh, the quantized 3D vertices are first partitioned into an octree structure, which is then traversed from the root and gradually to the leaves. During the traversal, each 3D cell in the tree front is subdivided into eight child cells. For each cell subdivision, both the local geometry and connectivity changes are encoded. Furthermore, selective cell subdivision is performed in the tree front to provide better rate-distortion performance. It is shown, by experimental results, that the geometry coding costs of the proposed codec are 4.2 and 14.5 bits per vertex (bpv) for 8-bit and 12-bit coordinate quantizations, respectively, and the connectivity coding cost is 3.3 bpv on the average, which provides state-of-the-art coding performance.