feat(nodes): add DepthAnythingDepthEstimationInvocation

Similar to the existing node, but without any resizing and with a revised model loading API.

invokeai/app/invocations/depth_anything.py

@@ -0,0 +1,45 @@
+from typing import Literal
+
+from invokeai.app.invocations.baseinvocation import BaseInvocation, invocation
+from invokeai.app.invocations.fields import ImageField, InputField, WithBoard, WithMetadata
+from invokeai.app.invocations.primitives import ImageOutput
+from invokeai.app.services.shared.invocation_context import InvocationContext
+from invokeai.backend.image_util.depth_anything.depth_anything_pipeline import DepthAnythingPipeline
+
+DEPTH_ANYTHING_MODEL_SIZES = Literal["large", "base", "small", "small_v2"]
+# DepthAnything V2 Small model is licensed under Apache 2.0 but not the base and large models.
+DEPTH_ANYTHING_MODELS = {
+    "large": "LiheYoung/depth-anything-large-hf",
+    "base": "LiheYoung/depth-anything-base-hf",
+    "small": "LiheYoung/depth-anything-small-hf",
+    "small_v2": "depth-anything/Depth-Anything-V2-Small-hf",
+}
+
+
+@invocation(
+    "depth_anything_depth_estimation",
+    title="Depth Anything Depth Estimation",
+    tags=["controlnet", "depth", "depth anything"],
+    category="controlnet",
+    version="1.0.0",
+)
+class DepthAnythingDepthEstimationInvocation(BaseInvocation, WithMetadata, WithBoard):
+    """Generates a depth map using a Depth Anything model."""
+
+    image: ImageField = InputField(description="The image to process")
+    model_size: DEPTH_ANYTHING_MODEL_SIZES = InputField(
+        default="small_v2", description="The size of the depth model to use"
+    )
+
+    def invoke(self, context: InvocationContext) -> ImageOutput:
+        model_url = DEPTH_ANYTHING_MODELS[self.model_size]
+        image = context.images.get_pil(self.image.image_name, "RGB")
+
+        loaded_model = context.models.load_remote_model(model_url, DepthAnythingPipeline.load_model)
+
+        with loaded_model as depth_anything_detector:
+            assert isinstance(depth_anything_detector, DepthAnythingPipeline)
+            depth_map = depth_anything_detector.generate_depth(image)
+
+        image_dto = context.images.save(image=depth_map)
+        return ImageOutput.build(image_dto)

invokeai/backend/image_util/depth_anything/depth_anything_pipeline.py

@@ -1,7 +1,9 @@
+import pathlib
 from typing import Optional
 
 import torch
 from PIL import Image
+from transformers import pipeline
 from transformers.pipelines import DepthEstimationPipeline
 
 from invokeai.backend.raw_model import RawModel
@@ -29,3 +31,11 @@ class DepthAnythingPipeline(RawModel):
         from invokeai.backend.model_manager.load.model_util import calc_module_size
 
         return calc_module_size(self._pipeline.model)
+
+    @classmethod
+    def load_model(cls, model_path: pathlib.Path):
+        """Load the model from the given path and return a DepthAnythingPipeline instance."""
+
+        depth_anything_pipeline = pipeline(model=str(model_path), task="depth-estimation", local_files_only=True)
+        assert isinstance(depth_anything_pipeline, DepthEstimationPipeline)
+        return cls(depth_anything_pipeline)