In this study, we investigate the role of mid-air gesture-based interaction technologies in cultural heritage learning. In an experiment, a mid-air gesture-based interactive media for Chinese Song Dynasty traditional painting-Ruihetu was developed and validated. Participants tested three experimental conditions: video learning only; interactive experience first, then video learning; and video learning first, then interactive experience. According to the research results, the outcomes of participants' learning of this cultural heritage differed significantly across all three experimental conditions. This study's findings offer insights into cultural learning of Chinese traditional painting in museums using mid-air gesture-based technology, specifically that video learning exhibits should be combined with and preceded by multimedia interactive exhibits for improved memory and understanding.