fix bug of first order multi person (#329)

ppgan/apps/first_order_predictor.py

@@ -33,6 +33,7 @@ from ppgan.faceutils import face_detection
 
 from .base_predictor import BasePredictor
 
+IMAGE_SIZE = 256
 
 class FirstOrderPredictor(BasePredictor):
     def __init__(self,
@@ -105,7 +106,6 @@ class FirstOrderPredictor(BasePredictor):
 
     def read_img(self, path):
         img = imageio.imread(path)
-        img = img.astype(np.float32)
         if img.ndim == 2:
             img = np.expand_dims(img, axis=2)
         # som images have 4 channels
@@ -161,14 +161,14 @@ class FirstOrderPredictor(BasePredictor):
         reader.close()
 
         driving_video = [
-            cv2.resize(frame, (256, 256)) / 255.0 for frame in driving_video
+            cv2.resize(frame, (IMAGE_SIZE, IMAGE_SIZE)) / 255.0 for frame in driving_video
         ]
         results = []
 
         # for single person
         if not self.multi_person:
             h, w, _ = source_image.shape
-            source_image = cv2.resize(source_image, (256, 256)) / 255.0
+            source_image = cv2.resize(source_image, (IMAGE_SIZE, IMAGE_SIZE)) / 255.0
             predictions = get_prediction(source_image)
             imageio.mimsave(os.path.join(self.output, self.filename), [
                 cv2.resize((frame * 255.0).astype('uint8'), (h, w))
@@ -181,7 +181,7 @@ class FirstOrderPredictor(BasePredictor):
         print(str(len(bboxes)) + " persons have been detected")
         if len(bboxes) <= 1:
             h, w, _ = source_image.shape
-            source_image = cv2.resize(source_image, (256, 256)) / 255.0
+            source_image = cv2.resize(source_image, (IMAGE_SIZE, IMAGE_SIZE)) / 255.0
             predictions = get_prediction(source_image)
             imageio.mimsave(os.path.join(self.output, self.filename), [
                 cv2.resize((frame * 255.0).astype('uint8'), (h, w))
@@ -193,7 +193,7 @@ class FirstOrderPredictor(BasePredictor):
         # for multi person
         for rec in bboxes:
             face_image = source_image.copy()[rec[1]:rec[3], rec[0]:rec[2]]
-            face_image = cv2.resize(face_image, (256, 256)) / 255.0
+            face_image = cv2.resize(face_image, (IMAGE_SIZE, IMAGE_SIZE)) / 255.0
             predictions = get_prediction(face_image)
             results.append({'rec': rec, 'predict': predictions})