3DCOMET: 3D compression methods test dataset

Abstract

HIGHLIGHTS• Dataset useful for testing 3D compression methods.• Contains more than 100 3D pointclouds with different texture and structure • Several tools for showing comparatives between methods.• Baseline information from state-of-the-arts compression methods. AbstractThe use of 3D data in mobile robotics applications provides valuable information about the robot's environment. However usually the huge amount of 3D information is difficult to manage due to the fact that the robot storage system and computing capabilities are insufficient. Therefore, a data compression method is necessary to store and process this information while preserving as much information as possible. A few methods have been proposed to compress 3D information. Nevertheless, there does not exist a consistent public benchmark for comparing the results (compression level, distance reconstructed error, etc.) obtained with different methods. In this paper, we propose a dataset composed of a set of 3D point clouds with different structure and texture variability to evaluate the results obtained from 3D data compression methods. We also provide useful tools for comparing compression methods, using as a baseline the results obtained by existing relevant compression methods.

Three Dimensional Point Cloud Compression and Decompression Using Polynomials of Degree One

“…Furthermore, high texture means more text available in the scene and vice versa. The above discussed data sets [8] are published in a well-known journal and these are publicly available under the license and further utilized in Miguel Cazorla compression methods [7,12]. They mainly concentrated on geometric characteristics of objects and scenes to compress point clouds.…”

“…They mainly concentrated on geometric characteristics of objects and scenes to compress point clouds. First, it utilized the advantageous aspects of Delaunay triangles to compress 3D surfaces in [7]. The point cloud is first segmented into a set of planes and then points under each plane are computed.…”

“…Furthermore, it achieved 55%-85% compression ratio with 0.5 RMSE. The compression technique proposed in [7] is further extended in [12] by the same author in which it utilized Gaussian Mixture Models (GMMs) and an Expectation-Maximization algorithm to compress point cloud data. The points of interest are selected as a first step in [12].…”

“…The points of interest are selected as a first step in [12]. These points are the best candidates to perform compression which are addressed by the planer patch extraction method as defined in [7]. Here GMMs are used to compress these points that hold the data as 3D multivariate Gaussian distribution, which allows the reconstruction of the original data with a certain statistical reliability.…”

“…The reconstructed data acquired from decompression phase is closely matching to the original one but not exactly. This compression results in loss of quality and reconstructed point cloud admit some errors [6,7]. So, the applications that utilize lossy compression techniques have to tolerate the difference between reconstructed and original data [8,9].…”