Added new correct cropping for training and smoothing for video.

natanielruiz

2017-08-08 18a21d4b07c581a8954b08518115fb035c712b28

 code/datasets.py

@@ -81,9 +81,24 @@
    def __getitem__(self, index):
        img = Image.open(os.path.join(self.data_dir, self.X_train[index] + self.img_ext))
        img = img.convert('RGB')
        mat_path = os.path.join(self.data_dir, self.y_train[index] + self.annot_ext)

        # Crop the face
        pt2d = utils.get_pt2d_from_mat(mat_path)
        x_min = min(pt2d[0,:])
        y_min = min(pt2d[1,:])
        x_max = max(pt2d[0,:])
        y_max = max(pt2d[1,:])

        k = 0.15
        x_min -= k * abs(x_max - x_min)
        y_min -= 4 * k * abs(y_max - y_min)
        x_max += k * abs(x_max - x_min)
        y_max += 0.4 * k * abs(y_max - y_min)
        img = img.crop((int(x_min), int(y_min), int(x_max), int(y_max)))

        # We get the pose in radians
        pose = utils.get_ypr_from_mat(os.path.join(self.data_dir, self.y_train[index] + self.annot_ext))
        pose = utils.get_ypr_from_mat(mat_path)
        # And convert to degrees.
        pitch = pose[0] * 180 / np.pi
        yaw = pose[1] * 180 / np.pi
@@ -117,23 +132,36 @@
    def __getitem__(self, index):
        img = Image.open(os.path.join(self.data_dir, self.X_train[index] + self.img_ext))
        img = img.convert('RGB')
        mat_path = os.path.join(self.data_dir, self.y_train[index] + self.annot_ext)

        # Crop the face
        pt2d = utils.get_pt2d_from_mat(mat_path)
        x_min = min(pt2d[0,:])
        y_min = min(pt2d[1,:])
        x_max = max(pt2d[0,:])
        y_max = max(pt2d[1,:])

        k = 0.15
        x_min -= k * abs(x_max - x_min)
        y_min -= 4 * k * abs(y_max - y_min)
        x_max += k * abs(x_max - x_min)
        y_max += 0.4 * k * abs(y_max - y_min)
        img = img.crop((int(x_min), int(y_min), int(x_max), int(y_max)))

        # We get the pose in radians
        pose = utils.get_ypr_from_mat(os.path.join(self.data_dir, self.y_train[index] + self.annot_ext))
        pose = utils.get_ypr_from_mat(mat_path)
        # And convert to degrees.
        pitch, yaw, roll = pose * 180 / np.pi
        pitch = pose[0] * 180 / np.pi
        yaw = pose[1] * 180 / np.pi
        roll = pose[2] * 180 / np.pi
        # Bin values
        bins = np.array(range(-99, 102, 3))
        binned_pitch = torch.DoubleTensor(np.digitize(pitch, bins) - 1)
        binned_yaw = torch.DoubleTensor(np.digitize(yaw, bins) - 1)
        binned_roll = torch.DoubleTensor(np.digitize(roll, bins) - 1)

        label = binned_yaw, binned_pitch, binned_roll
        labels = torch.LongTensor(np.digitize([yaw, pitch, roll], bins) - 1)

        if self.transform is not None:
            img = self.transform(img)

        return img, label, self.X_train[index]
        return img, labels, self.X_train[index]

    def __len__(self):
        # 2,000