Tool-IQA: Augmenting Image Quality Assessment with Simple Tools

Tool-IQA introduces a novel approach to Image Quality Assessment (IQA) by integrating tool-augmented workflows within Vision-Language Models (VLMs). It employs specific tools like a Magnifier for local detail inspection and a Gamma Corrector for enhancing visibility, alongside a batch-aware training strategy that optimizes tool usage. This method significantly improves IQA performance, achieving a Pearson Linear Correlation Coefficient (PLCC) of 0.854 on the CLIVE dataset, thus providing practitioners with a more dynamic and detailed assessment framework compared to traditional static scoring methods.