[fbsync] prevent unwrapping in SanitizeBoundingBoxes (#7446)

Reviewed By: vmoens

Differential Revision: D44416561

fbshipit-source-id: 8a782227bee116efaed2197365e8ccc30787ec70

Author: NicolasHug

test/test_transforms_v2.py

@@ -2020,6 +2020,9 @@ def test_sanitize_bounding_boxes(min_size, labels_getter, sample_type):
     assert out_image is input_img
     assert out_whatever is whatever
 
+    assert isinstance(out_boxes, datapoints.BoundingBox)
+    assert isinstance(out_masks, datapoints.Mask)
+
     if labels_getter is None or (callable(labels_getter) and labels_getter({"labels": "blah"}) is None):
         assert out_labels is labels
     else:

torchvision/transforms/v2/_misc.py

@@ -397,10 +397,15 @@ def forward(self, *inputs: Any) -> Any:
         return tree_unflatten(flat_outputs, spec)
 
     def _transform(self, inpt: Any, params: Dict[str, Any]) -> Any:
+        is_label = inpt is not None and inpt is params["labels"]
+        is_bounding_box_or_mask = isinstance(inpt, (datapoints.BoundingBox, datapoints.Mask))
 
-        if (inpt is not None and inpt is params["labels"]) or isinstance(
-            inpt, (datapoints.BoundingBox, datapoints.Mask)
-        ):
-            inpt = inpt[params["valid"]]
+        if not (is_label or is_bounding_box_or_mask):
+            return inpt
 
-        return inpt
+        output = inpt[params["valid"]]
+
+        if is_label:
+            return output
+
+        return type(inpt).wrap_like(inpt, output)