diff --git a/src/transformers/image_processing_utils.py b/src/transformers/image_processing_utils.py
index d3086de1cce4..52b798c09f84 100644
--- a/src/transformers/image_processing_utils.py
+++ b/src/transformers/image_processing_utils.py
@@ -39,6 +39,13 @@ class BaseImageProcessor(ImageProcessingMixin):
     def __init__(self, **kwargs):
         super().__init__(**kwargs)
 
+    @property
+    def is_fast(self) -> bool:
+        """
+        `bool`: Whether or not this image processor is a fast processor (backed by PyTorch and TorchVision).
+        """
+        return False
+
     def __call__(self, images, **kwargs) -> BatchFeature:
         """Preprocess an image or a batch of images."""
         return self.preprocess(images, **kwargs)
diff --git a/src/transformers/image_processing_utils_fast.py b/src/transformers/image_processing_utils_fast.py
index c575770a65f2..4fe9a48cb36f 100644
--- a/src/transformers/image_processing_utils_fast.py
+++ b/src/transformers/image_processing_utils_fast.py
@@ -235,6 +235,13 @@ def __init__(
         # get valid kwargs names
         self._valid_kwargs_names = list(self.valid_kwargs.__annotations__.keys())
 
+    @property
+    def is_fast(self) -> bool:
+        """
+        `bool`: Whether or not this image processor is a fast processor (backed by PyTorch and TorchVision).
+        """
+        return True
+
     def resize(
         self,
         image: "torch.Tensor",
diff --git a/tests/test_image_processing_common.py b/tests/test_image_processing_common.py
index 8f159045e68c..c8b75c73d984 100644
--- a/tests/test_image_processing_common.py
+++ b/tests/test_image_processing_common.py
@@ -36,7 +36,11 @@
     slow,
     torch_device,
 )
-from transformers.utils import is_torch_available, is_vision_available
+from transformers.utils import is_torch_available, is_torchvision_available, is_vision_available
+
+
+if is_torchvision_available():
+    from transformers.image_processing_utils_fast import BaseImageProcessorFast
 
 
 if is_torch_available():
@@ -241,6 +245,16 @@ def measure_time(image_processor, image):
 
         self.assertLessEqual(fast_time, slow_time)
 
+    def test_is_fast(self):
+        for image_processing_class in self.image_processor_list:
+            image_processor = image_processing_class(**self.image_processor_dict)
+
+            # Check is_fast is set correctly
+            if is_torchvision_available() and issubclass(image_processing_class, BaseImageProcessorFast):
+                self.assertTrue(image_processor.is_fast)
+            else:
+                self.assertFalse(image_processor.is_fast)
+
     def test_image_processor_to_json_string(self):
         for image_processing_class in self.image_processor_list:
             image_processor = image_processing_class(**self.image_processor_dict)