In-Vehicle Infotainment System GUI Video Dataset

With the rapid development of smart devices, in-vehicle infotainment systems face challenges in optimizing user experience and increasing interaction complexity. Currently, existing solutions show limitations in adapting to diverse user needs and complex operating scenarios. This dataset aims to solve the problems of automated analysis in in-vehicle system UI design and user behavior recognition, improving system recognition accuracy and user satisfaction. Data collection is conducted with high-definition cameras in real vehicle environments, covering a variety of vehicle models and interaction scenarios. Data quality is ensured through multiple rounds of annotation and consistency checks by expert review, and the annotation team consists of 20 members with professional backgrounds in human-computer interaction. Preprocessing steps include video clipping, image enhancement, and feature extraction, with the data format as MP4 organized by vehicle model and operation category. Core advantages include annotation accuracy of over 95% and 90% consistency, comprehensively covering various interaction modes of in-vehicle system interfaces. Through innovative video data enhancement technology and behavior recognition methods, performance improvement in in-vehicle system UI optimization of up to 20% is achieved. Compared to similar datasets, it offers more comprehensive data coverage and targeted analysis capabilities, with unique features such as diverse interaction scenarios and high-precision annotation, making it highly valuable for application and expansion in smart interaction research.