US 9,811,555 B2
Recognition of free-form gestures from orientation tracking of a handheld or wearable device
Nicholas G. Mitri, Hillsboro, OR (US); Christopher B. Wilkerson, Portland, OR (US); and Mariette Awad, Hillsboro, OR (US)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Sep. 27, 2014, as Appl. No. 14/499,133.
Prior Publication US 2016/0092504 A1, Mar. 31, 2016
Int. Cl. G06F 17/30 (2006.01); G06F 3/01 (2006.01); G06F 3/0487 (2013.01); G06F 1/16 (2006.01)
CPC G06F 17/30424 (2013.01) [G06F 1/163 (2013.01); G06F 1/1694 (2013.01); G06F 3/011 (2013.01); G06F 3/017 (2013.01); G06F 3/0487 (2013.01); G06F 17/30312 (2013.01); G06F 17/30598 (2013.01)] 25 Claims
OG exemplary drawing
 
1. An apparatus, comprising:
hardware logic configurable to cause a computing device to perform actions, the actions comprising:
capturing a sequence of rotation vectors from a device being moved to perform a gesture;
converting the sequence of rotation vectors to a sequence of corresponding device positions using a robotic chain model with at most 4 degrees of freedom; and
wherein the robotic chain model simulates a shoulder and an elbow of a user holding the device while performing the gesture; and
connecting the corresponding device positions to form a trace,
wherein the trace approximates a shape of the gesture; and
extracting features of the trace and using the features to classify the gesture;
wherein the gesture is classified by an algorithm comprising a support vector machine comparing the trace to contents of a training database and performing statistical analysis.