Efficient Contrast Effect Compensation with Personalized Perception Models

摘要: Color is one of the most effective visual variables and frequently used to encode metric quantities. Contrast effects are considered harmful in data visualizations since they significantly bias our perception colors. For instance, a gray patch appears brighter on black background than white background. Accordingly, color-encoded items depends surround rendered visualization. A method that compensates for contrast has been presented previously, which improves users' accuracy reading comparing color encoded data. The utilizes established models compensate effects, assuming an average human observer. In this paper, we provide experiments show significant difference users. We introduce methods personalize effect compensation outperforms original with user study. We, further, overcome major limitation method, runtime several minutes. With use efficient optimization surrogate models, able reduce milliseconds, making applicable interactive visualizations.