keep the resize function the same in test time the same with training time

demo/predictor.py

@@ -10,7 +10,38 @@
 from maskrcnn_benchmark import layers as L
 from maskrcnn_benchmark.utils import cv2_util
 
+class Resize(object):
+    def __init__(self, min_size, max_size):
+        self.min_size = min_size
+        self.max_size = max_size
+
+    # modified from torchvision to add support for max size
+    def get_size(self, image_size):
+        w, h = image_size
+        size = self.min_size
+        max_size = self.max_size
+        if max_size is not None:
+            min_original_size = float(min((w, h)))
+            max_original_size = float(max((w, h)))
+            if max_original_size / min_original_size * size > max_size:
+                size = int(round(max_size * min_original_size / max_original_size))
+
+        if (w <= h and w == size) or (h <= w and h == size):
+            return (h, w)
+
+        if w < h:
+            ow = size
+            oh = int(size * h / w)
+        else:
+            oh = size
+            ow = int(size * w / h)
+
+        return (oh, ow)
 
+    def __call__(self, image):
+        size = self.get_size(image.size)
+        image = F.resize(image, size)
+        return image
 class COCODemo(object):
     # COCO categories for pretty print
     CATEGORIES = [
@@ -147,11 +178,12 @@ def build_transform(self):
         normalize_transform = T.Normalize(
             mean=cfg.INPUT.PIXEL_MEAN, std=cfg.INPUT.PIXEL_STD
         )
-
+        min_size = cfg.INPUT.MIN_SIZE_TEST
+        max_size = cfg.INPUT.MAX_SIZE_TEST
         transform = T.Compose(
             [
                 T.ToPILImage(),
-                T.Resize(self.min_image_size),
+                Resize(min_size, max_size),
                 T.ToTensor(),
                 to_bgr_transform,
                 normalize_transform,