Wavelet Based Approach to 3D Geometry Point Cloud Compression
Abstract
Proposed is a method of 3D point cloud geometry compression. Point clouds find applications in autonomous driving, education, and other fields. The approach proposed here hopes to improve on existing quantization techniques by segmenting each coordinate axis into continuous regions of arbitrary size and preforming a wavelet transformation on these regions. This Thesis describes a means of finding these regions, taking the transform, quantizing, concatenating acquired data in a bitstream, recovering data from the bitstream, and reconstructing the point cloud. Results presented on solid, dense, and sparse point clouds show improvements relative to Quantization-Inverse Quantization at low bitrates but exhibit poor performance at higher bitrates and lower densities. A better sorting approach is expected to increase performance, but the need for continuous regions, and possible latency incurred when sorting for them imposes higher quality requirements on this approach.
Citation
Lesser, Max Michael Jakob (2023). Wavelet Based Approach to 3D Geometry Point Cloud Compression. Master's thesis, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /199145.