Intelligent Analysis and Classification of Piano Music Gestures with Multimodal Recordings

Table 2

Testing and training frequency analysis result for piano music gestures with multi-mode (image, audio/video and image).

Process	Data format	Mean	Standard deviation	Size (GB)
Testing	Image	1.54	275	65.87
	Video/Audio	2.89	365	72.64
	Text	1.65	245	64.23
Training	Image	1.67	162	39.46
	Video/Audio	1.89	203	42.12
	Text	1.23	153	32.46
Frequency	Image	1.14	242	48.69
	Video/Audio	2.12	239	47.23
	Text	1.89	212	34.12