Motivation System Design Experimnonts Coodusioo 急京大星 Finger Tapping Detection Audio to detect that the "tapping state" High sampling rate (48 kHz)->Low latency Sensitive in the depth direction->High accuracy High false positive rates Video to look back to the previous frames Measure the duration of "Moving state"and "Locating state" Check the state machine to remove false alarms->High robustness Measure the depth of finger tapping Keystroke localization Calculate the location of the fingertip during the "Locating state" Determine the fingertip with the largest bending angle 1-NN determine the pressed virtual key ·口⑨，4至·4子·2刀00 Ke Sun,et al Mobisys'18 June 3.2018 Depth Aware Finger Tapping on Virtual Display 13/22. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Motivation System Design Experiments Conclusion Finger Tapping Detection 1 Audio to detect that the ”tapping state” High sampling rate (48 kHz) –> Low latency Sensitive in the depth direction –> High accuracy High false positive rates 2 Video to look back to the previous frames Measure the duration of ”Moving state” and ”Locating state” Check the state machine to remove false alarms–> High robustness Measure the depth of finger tapping 3 Keystroke localization Calculate the location of the fingertip during the ”Locating state” Determine the fingertip with the largest bending angle 1-NN determine the pressed virtual key Ke Sun, et al. Mobisys’18 June 3, 2018 Depth Aware Finger Tapping on Virtual Display 13 / 22