The Worse The Better: Content-Aware Viewpoint Generation Network for Projection-related Point Cloud Quality Assessment

TL;DR

This paper introduces a content-aware viewpoint generation network that improves point cloud quality assessment by learning optimal viewpoints based on geometric and attribute features, leading to more stable quality scores.

Contribution

The paper proposes a novel content-aware viewpoint generation network (CAVGN) and a self-supervised ranking network (SSVRN) to enhance projection-based point cloud quality assessment.

Findings

Improved stability of quality scores across viewpoints.

Enhanced performance of PCQA methods with generated viewpoints.

Effective learning of viewpoints considering geometric and attribute features.

Abstract

Through experimental studies, however, we observed the instability of final predicted quality scores, which change significantly over different viewpoint settings. Inspired by the "wooden barrel theory", given the default content-independent viewpoints of existing projection-related PCQA approaches, this paper presents a novel content-aware viewpoint generation network (CAVGN) to learn better viewpoints by taking the distribution of geometric and attribute features of degraded point clouds into consideration. Firstly, the proposed CAVGN extracts multi-scale geometric and texture features of the entire input point cloud, respectively. Then, for each default content-independent viewpoint, the extracted geometric and texture features are refined to focus on its corresponding visible part of the input point cloud. Finally, the refined geometric and texture features are concatenated to generate an optimized viewpoint. To train the proposed CAVGN, we present a self-supervised viewpoint ranking network (SSVRN) to select the viewpoint with the worst quality projected image to construct a default-optimized viewpoint dataset, which consists of thousands of paired default viewpoints and corresponding optimized viewpoints. Experimental results show that the projection-related PCQA methods can achieve higher performance using the viewpoints generated by the proposed CAVGN.