Open HimGautam opened 3 years ago
In the code the disparity map is having non linearity as given below,
def get_disp(self, x): disp = 0.3 * self.conv(x, 2, 3, 1, tf.nn.sigmoid) return disp
whereas in the paper non linearity of disparity map is given by scaled sigmoid (0, dmax), where dmax = 0.3× the image width at a given output scale. Which one should I use?
In the code the disparity map is having non linearity as given below,
whereas in the paper non linearity of disparity map is given by scaled sigmoid (0, dmax), where dmax = 0.3× the image width at a given output scale. Which one should I use?