Add text to vision embedding

monai/networks/blocks/text_embedding.py

@@ -0,0 +1,85 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import torch
+from torch import nn
+from torch.utils import model_zoo
+
+url_map = {
+    "clip_encoding_univeral_model_32": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/clip_encoding_univeral_model.pth",
+}
+
+
+class TextEncoder(nn.Module):
+    """
+    Text to vision encoding by Contrastive Language-Image Pre-training (CLIP) or random embedding.
+    The text to vision encoder loads the pre-trained or random initialized weights with connection to 2D/3D vision models.
+
+    Contrastive Language-Image Pre-training (CLIP), based on: "Radford et al.,
+    Learning Transferable Visual Models From Natural Language Supervision <https://arxiv.org/abs/2103.00020>"
+
+    Connecting text and medical 3D image, based on: "Liu et al.,
+    CLIP-Driven Universal Model for Organ Segmentation and Tumor Detection <https://arxiv.org/pdf/2301.00785.pdf>"
+    """
+    def __init__(
+        self,
+        out_channels: int,
+        spatial_dims: int = 3,
+        text_dim: int = 512,
+        hidden_size: int = 256,
+        encoding: str = "rand_embedding",
+        pretrained: bool = False
+    ) -> None:
+        """
+        Args:
+            out_channels: number of output channels, to control text-baesd embedding for classes.
+            spatial_dims: number of spatial dims.
+            text_dim: dimension of text embeddings.
+            hidden_size: dimension of hidden features, compatible to different vision feature dimensions.
+            encoding: the text embedding type, default to use clip text pretrained weights.
+            pretrained: whether to load pretrained weights from e.g., (CLIP) to initialize text embeddings, default to False.
+        """
+        super().__init__()
+        self.encoding = encoding
+
+        self.spatial_dims = spatial_dims
+        if spatial_dims not in (2, 3):
+            raise ValueError("spatial dimension should be 2 or 3.")
+
+        if self.encoding == 'rand_embedding':
+            self.text_embedding = nn.Embedding(out_channels, hidden_size)
+        else:
+            self.register_buffer('text_embedding', torch.randn(out_channels, text_dim))
+
+            if pretrained:
+                model_url = url_map[self.encoding]
+                pretrain_state_dict = model_zoo.load_url(model_url)
+                self.text_embedding.data = pretrain_state_dict.float()
+            else:
+                print(f'{self.encoding} is not implemented, and can not be downloaded, please load your own')
+
+            self.text_to_vision = nn.Linear(text_dim, hidden_size)
+
+    def forward(self):
+        if self.encoding == 'rand_embedding':
+            # text embedding as random initialized 'rand_embedding'
+            test_encoding = self.text_embedding.weight
+        else:
+            test_encoding = nn.functional.relu(self.text_to_vision(self.text_embedding))
+
+        if self.spatial_dims == 3:
+            test_encoding = test_encoding.unsqueeze(2).unsqueeze(2).unsqueeze(2)
+        elif self.spatial_dims == 2:
+            test_encoding = test_encoding.unsqueeze(2).unsqueeze(2)
+
+        return test_encoding

tests/test_text_encoding.py

@@ -0,0 +1,47 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from monai.networks.blocks.text_embedding import TextEncoder
+from tests.utils import skip_if_downloading_fails
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+
+class TestTextEncoder(unittest.TestCase):
+    def test_test_encoding_shape(self):
+        with skip_if_downloading_fails():
+            # test 2D encoder
+            text_encoder = TextEncoder(spatial_dims=2, out_channels=32, encoding="clip_encoding_univeral_model_32", pretrained=True).to(device)
+            text_encoding = text_encoder()
+            self.assertEqual(text_encoding.shape, (32,256,1,1))
+
+            # test 3D encoder
+            text_encoder = TextEncoder(spatial_dims=3, out_channels=32, encoding="clip_encoding_univeral_model_32", pretrained=True).to(device)
+            text_encoding = text_encoder()
+            self.assertEqual(text_encoding.shape, (32,256,1,1,1))
+
+        # test random enbedding 3D
+        text_encoder = TextEncoder(spatial_dims=3, out_channels=32, encoding="rand_embedding", pretrained=True).to(device)
+        text_encoding = text_encoder()
+        self.assertEqual(text_encoding.shape, (32,256,1,1,1))
+
+        # test random enbedding 2D
+        text_encoder = TextEncoder(spatial_dims=2, out_channels=32, encoding="rand_embedding", pretrained=True).to(device)
+        text_encoding = text_encoder()
+        self.assertEqual(text_encoding.shape, (32,256,1,1))
+
+if __name__ == "__main__":
+    unittest.main()