ColorBench: Can VLMs See and Understand the Colorful World? A Comprehensive Benchmark for Color Perception, Reasoning, and Robustness

Ming Li (University of Central Florida) · Tianyi Zhou (University of Southern California) · Yijun Liang (University of Maryland, College Park) · Chenrui Fan (University of Maryland, College Park) · Ziyue Li (Technical University of Munich (TUM)) · Dang Nguyen (University of Maryland, College Park) · Kwesi Cobbina (University of Maryland, College Park) · Shweta Bhardwaj (University of Maryland College Park) · Jiuhai Chen (University of Maryland) · Fuxiao Liu (NVIDIA)

color perceptioncolor transformationscolor understandingcolor-based cuescolorbenchcot reasoninghuman-level understandinglanguage modelmultimodal aiperformance gaps robustness scaling lawtest scenariosvision encoder vision-language models visual reasoning

Color plays an important role in human perception and usually provides critical clues in visual reasoning. However, it is unclear whether and how vision-language models (VLMs) can perceive, understand, and leverage color as humans.This paper introduces ColorBench, an innovative benchmark meticulously crafted to assess the capabilities of VLMs in color understanding, including color perception, reasoning, and robustness. By curating a suite of diverse test scenarios, with grounding in real applications, ColorBench evaluates how these models perceive colors, infer meanings from color-based cues, and maintain consistent performance under varying color transformations. Through an extensive evaluation of 32 VLMs with varying language models and vision encoders, our paper reveals some undiscovered findings: (i) The scaling law (larger models are better) still holds on ColorBench, while the language model plays a more important role than the vision encoder. (ii) However, the performance gaps across models are relatively small, indicating that color understanding has been largely neglected by existing VLMs. (iii) CoT reasoning improves color understanding accuracies and robustness, though they are vision-centric tasks. (iv) Color clues are indeed leveraged by VLMs on ColorBench but they can also mislead models in some tasks.These findings highlight the critical limitations of current VLMs and underscore the need to enhance color comprehension. Our ColorBench can serve as a foundational tool for advancing the study of human-level color understanding of multimodal AI.