Use pip to install OpenMMLab dependencies

invokeai/app/invocations/controlnet_image_processors.py

@@ -6,9 +6,11 @@ from typing import Dict, List, Literal, Optional, Union
 
 import cv2
 import numpy as np
+import torch
 from controlnet_aux import (
     CannyDetector,
     ContentShuffleDetector,
+    DWposeDetector,
     HEDdetector,
     LeresDetector,
     LineartAnimeDetector,
@@ -125,7 +127,7 @@ class ControlNetInvocation(BaseInvocation):
 
 
 @invocation(
-    "image_processor", title="Base Image Processor", tags=["controlnet"], category="controlnet", version="1.0.0"
+    "image_processor", title="Base Image Processorwp", tags=["controlnet"], category="controlnet", version="1.0.0"
 )
 class ImageProcessorInvocation(BaseInvocation):
     """Base class for invocations that preprocess images for ControlNet"""
@@ -589,3 +591,29 @@ class ColorMapImageProcessorInvocation(ImageProcessorInvocation):
         color_map = cv2.resize(color_map, (width, height), interpolation=cv2.INTER_NEAREST)
         color_map = Image.fromarray(color_map)
         return color_map
+
+
+@invocation(
+    "dwpose_image_processor",
+    title="DWPose Processor",
+    tags=["controlnet", "dwpose", "pose"],
+    category="controlnet",
+    version="1.0.0",
+)
+class DWPoseImageProcessorInvocation(ImageProcessorInvocation):
+    """Applies DW-Pose processing to image"""
+
+    detect_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.detect_res)
+    image_resolution: int = InputField(default=512, ge=0, description=FieldDescriptions.image_res)
+
+    def run_processor(self, image):
+        # device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+        # for now, executing DWPose processing on CPU only
+        device = "cpu"
+        dwpose_processor = DWposeDetector(device=device)
+        processed_image = dwpose_processor(
+            image,
+            detect_resolution=self.detect_resolution,
+            image_resolution=self.image_resolution,
+        )
+        return processed_image

pyproject.toml

@@ -37,7 +37,7 @@ dependencies = [
   "click",
   "clip_anytorch",  # replacing "clip @ https://github.com/openai/CLIP/archive/eaa22acb90a5876642d0507623e859909230a52d.zip",
   "compel~=2.0.2",
-  "controlnet-aux>=0.0.6",
+  "controlnet-aux>=0.0.7",
   "timm==0.6.13",   # needed to override timm latest in controlnet_aux, see  https://github.com/isl-org/ZoeDepth/issues/26
   "datasets",
   "diffusers[torch]~=0.21.0",
@@ -52,6 +52,10 @@ dependencies = [
   "invisible-watermark~=0.2.0", # needed to install SDXL base and refiner using their repo_ids
   "matplotlib",                 # needed for plotting of Penner easing functions
   "mediapipe",                  # needed for "mediapipeface" controlnet model
+  "mmcv>=2.0.1",
+  "mmdet>=3.1.0",
+  "mmengine",
+  "mmpose>=1.1.0",
   "numpy",
   "npyscreen",
   "omegaconf",