Thanks for your excellent work. I use the pdb.m script the compute the distribution of the pose coefficients of the 300W fullset. But the posePara is between 0.1184 and 776.2491(I choose the first dim of the COEFF as the poseEig), this result looks weird because the result on the AFLW is between 0.0028 and 94.8463. And I want to know why the value of x-axis in Figure5 shown in your paper is between -80 and 80. Waiting for your reply。
Thanks for your excellent work. I use the pdb.m script the compute the distribution of the pose coefficients of the 300W fullset. But the posePara is between 0.1184 and 776.2491(I choose the first dim of the COEFF as the poseEig), this result looks weird because the result on the AFLW is between 0.0028 and 94.8463. And I want to know why the value of x-axis in Figure5 shown in your paper is between -80 and 80. Waiting for your reply。