ColorFoil: Investigating Color Blindness in Large Vision and Language Models
This work addresses the problem of color blindness in AI models for researchers and developers, but it is incremental as it focuses on benchmarking existing models without proposing new methods.
The paper introduced ColorFoil, a benchmark to assess color perception in large vision and language models, finding that ViLT and BridgeTower perform better than CLIP-based models and GroupViT, which struggle with distinguishing visually distinct colors.
With the utilization of Transformer architecture, large Vision and Language (V&L) models have shown promising performance in even zero-shot settings. Several studies, however, indicate a lack of robustness of the models when dealing with complex linguistics and visual attributes. In this work, we introduce a novel V&L benchmark - ColorFoil, by creating color-related foils to assess the models' perception ability to detect colors like red, white, green, etc. We evaluate seven state-of-the-art V&L models including CLIP, ViLT, GroupViT, and BridgeTower, etc. in a zero-shot setting and present intriguing findings from the V&L models. The experimental evaluation indicates that ViLT and BridgeTower demonstrate much better color perception capabilities compared to CLIP and its variants and GroupViT. Moreover, CLIP-based models and GroupViT struggle to distinguish colors that are visually distinct to humans with normal color perception ability.