diff --git a/src/diffusers/image_processor.py b/src/diffusers/image_processor.py
index 4f6f4bd52b9c..42339a2baccf 100644
--- a/src/diffusers/image_processor.py
+++ b/src/diffusers/image_processor.py
@@ -885,7 +885,7 @@ def preprocess(
         height: int | None = None,
         width: int | None = None,
         padding_mask_crop: int | None = None,
-    ) -> tuple[torch.Tensor, torch.Tensor]:
+    ) -> tuple[torch.Tensor, torch.Tensor | None, dict]:
         """
         Preprocess the image and mask.
         """
@@ -894,7 +894,12 @@ def preprocess(
 
         # if mask is None, same behavior as regular image processor
         if mask is None:
-            return self._image_processor.preprocess(image, height=height, width=width)
+            processed_image = self._image_processor.preprocess(image, height=height, width=width)
+            return processed_image, None, {
+                "crops_coords": None,
+                "original_image": None,
+                "original_mask": None,
+            }
 
         if padding_mask_crop is not None:
             crops_coords = self._image_processor.get_crop_region(mask, width, height, pad=padding_mask_crop)
diff --git a/tests/others/test_image_processor.py b/tests/others/test_image_processor.py
index 0d358699f105..487c32299a79 100644
--- a/tests/others/test_image_processor.py
+++ b/tests/others/test_image_processor.py
@@ -17,7 +17,7 @@
 import PIL.Image
 import torch
 
-from diffusers.image_processor import VaeImageProcessor
+from diffusers.image_processor import InpaintProcessor, VaeImageProcessor
 
 
 class TestImageProcessor:
@@ -306,3 +306,17 @@ def test_vae_image_processor_resize_np(self):
         assert out_np.shape == exp_np_shape, (
             f"resized image output shape '{out_np.shape}' didn't match expected shape '{exp_np_shape}'."
         )
+
+    def test_inpaint_processor_without_mask_returns_three_values(self):
+        processor = InpaintProcessor(do_resize=False)
+        image = PIL.Image.fromarray(np.zeros((8, 8, 3), dtype=np.uint8))
+
+        processed_image, processed_mask, postprocessing_kwargs = processor.preprocess(image)
+
+        assert processed_image.shape == (1, 3, 8, 8)
+        assert processed_mask is None
+        assert postprocessing_kwargs == {
+            "crops_coords": None,
+            "original_image": None,
+            "original_mask": None,
+        }