Video-Based Dynamic Mesh Coding

Abstract

In this paper we present a video-based approach for coding dynamic meshes, e.g. meshes that change their connectivity at each frame. Inspired by the new Visual Volumetric Video-Based Coding (V3C) standard, which is used to code volumetric content such as point clouds and immersive video, we present a method to encode meshes by using orthogonal projections, followed by atlas packing and video coding. The approach leverages the widely available video codecs and allows the reconstruction of high-quality dynamic meshes from videos and a thin layer of metadata information. It extends the capability of the V3C standards to include dynamic meshes. We present experimental results with dynamic meshes converted from high-quality point clouds that were used during the point cloud standardization process. Additionally, we use a metric that is being considered by the MPEG group for mesh evaluation and compare our results with a state-of-the-art mesh compression approach.