R4-CGQA: Retrieval-based Vision Language Models for Computer Graphics Image Quality Assessment

TL;DR

This paper introduces R4-CGQA, a retrieval-augmented vision language model framework that improves computer graphics image quality assessment by leveraging a new dataset and question-answer benchmarks, addressing the limitations of existing methods.

Contribution

The paper constructs a new dataset with quality descriptions for CG images and develops a retrieval-based framework to enhance VLMs' ability to assess CG quality accurately.

Findings

Current VLMs struggle with fine-grained CG quality judgment.

Retrieval-augmented generation significantly improves assessment performance.

Descriptions of similar images boost VLM understanding of CG quality.

Abstract

Immersive Computer Graphics (CGs) rendering has become ubiquitous in modern daily life. However, comprehensively evaluating CG quality remains challenging for two reasons: First, existing CG datasets lack systematic descriptions of rendering quality; and second existing CG quality assessment methods cannot provide reasonable text-based explanations. To address these issues, we first identify six key perceptual dimensions of CG quality from the user perspective and construct a dataset of 3500 CG images with corresponding quality descriptions. Each description covers CG style, content, and perceived quality along the selected dimensions. Furthermore, we use a subset of the dataset to build several question-answer benchmarks based on the descriptions in order to evaluate the responses of existing Vision Language Models (VLMs). We find that current VLMs are not sufficiently accurate in judging fine-grained CG quality, but that descriptions of visually similar images can significantly improve a VLM's understanding of a given CG image. Motivated by this observation, we adopt retrieval-augmented generation and propose a two-stream retrieval framework that effectively enhances the CG quality assessment capabilities of VLMs. Experiments on several representative VLMs demonstrate that our method substantially improves their performance on CG quality assessment.