I just need help with #8, computing the matrices for each image patch, thanks!

4 Great Features to Track [Opts] In several places we have seen this matrix, which is the summation over some window (area) in the image of the intensity gradients: U, V u. V 2u u For determining which parts of an image are great features for tracking, we want to detect good Corner-ness. Some metrics discussed include C1 which is the ratio of the eigen values of H, /X2. Simpler functions to compute include: 4 Great Features to Track [Opts] In several places we have seen this matrix, which is the summation over some window (area) in the image of the intensity gradients: U, V u. V 2u u For determining which parts of an image are great features for tracking, we want to detect good Corner-ness. Some metrics discussed include C1 which is the ratio of the eigen values of H, /X2. Simpler functions to compute include