Example-based super-resolution for point-cloud video

TL;DR

This paper introduces an example-based super-resolution method for point-cloud videos that enhances resolution by leveraging similarities between frames, improving quality and enabling various processing tools.

Contribution

It presents a novel mixed-resolution point-cloud representation and super-resolution framework that infers high-frequency details from adjacent frames, advancing point-cloud processing techniques.

Findings

Achieves an average 1.18 dB gain over low-pass point-clouds

Enables effective compression, denoising, and error concealment

Improves high-frequency content inference from frame similarities

Abstract

We propose a mixed-resolution point-cloud representation and an example-based super-resolution framework, from which several processing tools can be derived, such as compression, denoising and error concealment. By inferring the high-frequency content of low-resolution frames based on the similarities between adjacent full-resolution frames, the proposed framework achieves an average 1.18 dB gain over low-pass versions of the point-cloud, for a projection-based distortion metric[1-2].