Learning Convolutional Transforms for Point Cloud Geometry Compression: Supplementary Material

[Paper] [Source code] [Supplementary Material Website] [Supplementary Material]

Select a sample and a corresponding lambda value to compare the results obtained using our method and using the MPEG anchor. Higher lambda values target higher bitrates while lower lambda values target lower bitrates.

Sample Lambda value:

Learning Convolutional Transforms for Point Cloud Geometry Compression: Supplementary Material

Original point cloud

Our method

MPEG anchor