Add tests for negative samples for Mask R-CNN and Keypoint R-CNN (#2069)

* Add tests for negative samples for Mask R-CNN and Keypoint R-CNN

* Fix lint

Author: fmassa

test/test_models_detection_negative_samples.py

@@ -11,16 +11,27 @@
 
 class Tester(unittest.TestCase):
 
-    def test_targets_to_anchors(self):
+    def _make_empty_sample(self, add_masks=False, add_keypoints=False):
+        images = [torch.rand((3, 100, 100), dtype=torch.float32)]
         boxes = torch.zeros((0, 4), dtype=torch.float32)
         negative_target = {"boxes": boxes,
-                           "labels": torch.zeros((1, 1), dtype=torch.int64),
+                           "labels": torch.zeros(0, dtype=torch.int64),
                            "image_id": 4,
                            "area": (boxes[:, 3] - boxes[:, 1]) * (boxes[:, 2] - boxes[:, 0]),
                            "iscrowd": torch.zeros((0,), dtype=torch.int64)}
 
-        anchors = [torch.randint(-50, 50, (3, 4), dtype=torch.float32)]
+        if add_masks:
+            negative_target["masks"] = torch.zeros(0, 100, 100, dtype=torch.uint8)
+
+        if add_keypoints:
+            negative_target["keypoints"] = torch.zeros(17, 0, 3, dtype=torch.float32)
+
         targets = [negative_target]
+        return images, targets
+
+    def test_targets_to_anchors(self):
+        _, targets = self._make_empty_sample()
+        anchors = [torch.randint(-50, 50, (3, 4), dtype=torch.float32)]
 
         anchor_sizes = ((32,), (64,), (128,), (256,), (512,))
         aspect_ratios = ((0.5, 1.0, 2.0),) * len(anchor_sizes)
@@ -85,25 +96,37 @@ def test_assign_targets_to_proposals(self):
         self.assertEqual(labels[0].shape, torch.Size([proposals[0].shape[0]]))
         self.assertEqual(labels[0].dtype, torch.int64)
 
-    def test_forward_negative_sample(self):
-        model = torchvision.models.detection.fasterrcnn_resnet50_fpn(pretrained=True)
-        in_features = model.roi_heads.box_predictor.cls_score.in_features
-        model.roi_heads.box_predictor = FastRCNNPredictor(in_features, 2)
+    def test_forward_negative_sample_frcnn(self):
+        model = torchvision.models.detection.fasterrcnn_resnet50_fpn(
+            num_classes=2, min_size=100, max_size=100)
 
-        images = [torch.rand((3, 100, 100), dtype=torch.float32)]
-        boxes = torch.zeros((0, 4), dtype=torch.float32)
-        negative_target = {"boxes": boxes,
-                           "labels": torch.zeros((1, 1), dtype=torch.int64),
-                           "image_id": 4,
-                           "area": (boxes[:, 3] - boxes[:, 1]) * (boxes[:, 2] - boxes[:, 0]),
-                           "iscrowd": torch.zeros((0,), dtype=torch.int64)}
+        images, targets = self._make_empty_sample()
+        loss_dict = model(images, targets)
 
-        targets = [negative_target]
+        self.assertEqual(loss_dict["loss_box_reg"], torch.tensor(0.))
+        self.assertEqual(loss_dict["loss_rpn_box_reg"], torch.tensor(0.))
+
+    def test_forward_negative_sample_mrcnn(self):
+        model = torchvision.models.detection.maskrcnn_resnet50_fpn(
+            num_classes=2, min_size=100, max_size=100)
+
+        images, targets = self._make_empty_sample(add_masks=True)
+        loss_dict = model(images, targets)
+
+        self.assertEqual(loss_dict["loss_box_reg"], torch.tensor(0.))
+        self.assertEqual(loss_dict["loss_rpn_box_reg"], torch.tensor(0.))
+        self.assertEqual(loss_dict["loss_mask"], torch.tensor(0.))
+
+    def test_forward_negative_sample_krcnn(self):
+        model = torchvision.models.detection.keypointrcnn_resnet50_fpn(
+            num_classes=2, min_size=100, max_size=100)
 
+        images, targets = self._make_empty_sample(add_keypoints=True)
         loss_dict = model(images, targets)
 
         self.assertEqual(loss_dict["loss_box_reg"], torch.tensor(0.))
         self.assertEqual(loss_dict["loss_rpn_box_reg"], torch.tensor(0.))
+        self.assertEqual(loss_dict["loss_keypoint"], torch.tensor(0.))
 
 
 if __name__ == '__main__':

torchvision/ops/misc.py

@@ -113,7 +113,7 @@ class can go away.
         )
 
     output_shape = _output_size(2, input, size, scale_factor)
-    output_shape = input.shape[:-2] + output_shape
+    output_shape = list(input.shape[:-2]) + output_shape
     return _new_empty_tensor(input, output_shape)