Multimodal Data Collection

The CMU Multi-Modal Activity Database (CMU-MMAC) database contains multimodal measures of the human activity of subjects performing the tasks involved in cooking and food preparation. The CMU-MMAC database was collected in Carnegie Mellon University’s Motion Capture Lab. A kitchen was built, and, to date, five subjects have been recorded cooking five different recipes: brownies, pizza, sandwich, salad and scrambled eggs. The following modalities were recorded:

• Video: (A) Three high spatial resolution (1024×768) color video cameras at low temporal resolution (30 Hz). (B) Two low spatial resolution (640×480) color video cameras at high temporal resolution (60 Hz). (C) One wearable low spatial resolution (640×480) camera at low temporal resolution (12 Hz).
• Audio: (A) Five balanced microphones. (B) Wearable watch.
• Motion capture: A Vicon motion capture system with 12 infrared MX-40 cameras. Each camera records 4 megapixel resolution images at 120 Hz.
• Five 3-axis accelerometers and gyroscopes.