GT-PCQA: Geometry-Texture Decoupled Point Cloud Quality Assessment with MLLM

TL;DR

This paper introduces GT-PCQA, a novel framework for point cloud quality assessment that decouples geometry and texture features, leveraging multi-modal large language models with a joint training strategy and dual-prompt mechanism.

Contribution

The paper proposes a geometry-texture decoupling strategy and a 2D-3D joint training approach to improve MLLM-based point cloud quality assessment under limited supervision.

Findings

GT-PCQA achieves competitive performance in PCQA tasks.

The dual-prompt mechanism enhances sensitivity to geometric degradations.

Joint training unifies IQA and PCQA datasets effectively.

Abstract

With the rapid advancement of Multi-modal Large Language Models (MLLMs), MLLM-based Image Quality Assessment (IQA) methods have shown promising generalization. However, directly extending these MLLM-based IQA methods to PCQA remains challenging. On the one hand, existing PCQA datasets are limited in scale, which hinders stable and effective instruction tuning of MLLMs. On the other hand, due to large-scale image-text pretraining, MLLMs tend to rely on texture-dominant reasoning and are insufficiently sensitive to geometric structural degradations that are critical for PCQA. To address these gaps, we propose a novel MLLM-based no-reference PCQA framework, termed GT-PCQA, which is built upon two key strategies. First, to enable stable and effective instruction tuning under scarce PCQA supervision, a 2D-3D joint training strategy is proposed. This strategy formulates PCQA as a relative quality comparison problem to unify large-scale IQA datasets with limited PCQA datasets. It incorporates a parameter-efficient Low-Rank Adaptation (LoRA) scheme to support instruction tuning. Second, a geometry-texture decoupling strategy is presented, which integrates a dual-prompt mechanism with an alternating optimization scheme to mitigate the inherent texture-dominant bias of pre-trained MLLMs, while enhancing sensitivity to geometric structural degradations. Extensive experiments demonstrate that GT-PCQA achieves competitive performance and exhibits strong generalization.