US 9,811,917 B2
Confidence estimation for optical flow under low light conditions
Manu Mathew, Bangalore (IN); Hrushikesh Tukaram Garud, Parbhani (IN); and Soyeb Noormohammed Nagori, Bangalore (IN)
Filed by Texas Instruments Incorporated, Dallas, TX (US)
Filed on May 31, 2016, as Appl. No. 15/169,153.
Claims priority of application No. 201641001331 (IN), filed on Jan. 14, 2016.
Prior Publication US 2017/0206665 A1, Jul. 20, 2017
Int. Cl. G06K 9/00 (2006.01); G06T 7/20 (2017.01); G06K 9/48 (2006.01); G06K 9/62 (2006.01); G06T 5/00 (2006.01)
CPC G06T 7/2066 (2013.01) [G06K 9/481 (2013.01); G06K 9/6282 (2013.01); G06T 5/002 (2013.01); G06T 7/2033 (2013.01)] 8 Claims
OG exemplary drawing
1. A method of confidence estimation for optical flow comprising the steps of:
computing the census texture gradient (Cg) features for each pixel of an input image;
computing smoothened versions of said features (Sg);
constructing a feature vector for each pixel of the input image;
computing a classifier score from said features by employing a decision tree classifier;
converting the classifier score into a confidence score.