Today’s Patent – Three-Dimensional Mesh Compression Using a Video Encoder

The said invention (US11783508B2) was invented by Khaled Mammou, Alexandros Tourapis, and Jungsun KIM, on 24^th September 2019. Currently; it stands assigned to Carl Zeiss Apple Inc.

The system features an encoder that compresses and encodes three-dimensional meshes using video encoding techniques. It identifies sub-meshes and their texture and geometry patches, along with connectivity information, and packs these into video frames for encoding. A decoder then reconstructs the mesh from the compressed bit stream, using the connectivity and stitching data to merge sub-meshes.

The system can capture spatial and attribute data through various sensors or generate meshes without prior point cloud modeling. The encoder computes depth values for vertices, while the decoder applies this information to reconstruct the original three-dimensional object, including interpolating texture or attribute values for accurate representation.