change control model

app_haircolor_15.py

@@ -0,0 +1,158 @@
+import spaces
+import gradio as gr
+import time
+import torch
+import numpy as np
+
+from PIL import Image
+from segment_utils import(
+    segment_image_withmask,
+    restore_result,
+)
+from diffusers import (
+    StableDiffusionControlNetImg2ImgPipeline,
+    ControlNetModel,
+    DDIMScheduler,
+)
+
+from controlnet_aux import (
+    CannyDetector,
+)
+
+BASE_MODEL = "stable-diffusion-v1-5/stable-diffusion-v1-5"
+
+DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+
+DEFAULT_EDIT_PROMPT = "a woman, blue hair, high detailed"
+DEFAULT_NEGATIVE_PROMPT = "worst quality, normal quality, low quality, low res, blurry, text, watermark, logo, banner, extra digits, cropped, jpeg artifacts, signature, username, error, sketch ,duplicate, ugly, monochrome, horror, geometry, mutation, disgusting, poorly drawn face, bad face, fused face, ugly face, worst face, asymmetrical, unrealistic skin texture, bad proportions, out of frame, poorly drawn hands, cloned face, double face"
+
+DEFAULT_CATEGORY = "hair"
+
+canny_detector = CannyDetector()
+
+controlnet = [
+    ControlNetModel.from_pretrained(
+        "lllyasviel/control_v11p_sd15_inpaint", 
+        torch_dtype=torch.float16,
+    ),
+    ControlNetModel.from_pretrained(
+        "diffusers/controlnet-canny-sdxl-1.0",
+        torch_dtype=torch.float16,
+    ),
+]
+
+basepipeline = StableDiffusionControlNetImg2ImgPipeline.from_pretrained(
+    BASE_MODEL,
+    torch_dtype=torch.float16,
+    use_safetensors=True,
+    controlnet=controlnet,
+)
+basepipeline.scheduler = DDIMScheduler.from_config(basepipeline.scheduler.config)
+
+basepipeline = basepipeline.to(DEVICE)
+
+basepipeline.enable_model_cpu_offload()
+
+@spaces.GPU(duration=30)
+def image_to_image(
+    input_image: Image,
+    mask_image: Image,
+    edit_prompt: str,
+    seed: int,
+    num_steps: int,
+    guidance_scale: float,
+    generate_size: int,
+    lineart_scale: float = 1.0,
+    hed_scale: float = 0.5,
+):
+    run_task_time = 0
+    time_cost_str = ''
+    run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
+    canny_image = canny_detector(input_image, int(generate_size*0.375), generate_size)
+    inpaint_image = make_inpaint_condition(input_image, mask_image)
+    control_image = [inpaint_image, canny_image]
+
+    generator = torch.Generator(device=DEVICE).manual_seed(seed)
+    generated_image = basepipeline(
+        generator=generator,
+        prompt=edit_prompt,
+        negative_prompt=DEFAULT_NEGATIVE_PROMPT,
+        image=input_image,
+        mask_image=mask_image,
+        control_image=control_image,
+        guidance_scale=guidance_scale,
+        num_inference_steps=num_steps,
+        controlnet_conditioning_scale=[inpaint_scale, canny_scale]
+    ).images[0]
+    
+    run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
+
+    return generated_image, time_cost_str
+
+def make_inpaint_condition(image, image_mask):
+    image = np.array(image.convert("RGB")).astype(np.float32) / 255.0
+    image_mask = np.array(image_mask.convert("L")).astype(np.float32) / 255.0
+
+    assert image.shape[0:1] == image_mask.shape[0:1], "image and image_mask must have the same image size"
+    image[image_mask > 0.5] = -1.0  # set as masked pixel
+    image = np.expand_dims(image, 0).transpose(0, 3, 1, 2)
+    image = torch.from_numpy(image)
+    return image
+
+def get_time_cost(run_task_time, time_cost_str):
+    now_time = int(time.time()*1000)
+    if run_task_time == 0:
+        time_cost_str = 'start'
+    else:
+        if time_cost_str != '': 
+            time_cost_str += f'-->'
+        time_cost_str += f'{now_time - run_task_time}'
+    run_task_time = now_time
+    return run_task_time, time_cost_str
+
+def create_demo() -> gr.Blocks:
+    with gr.Blocks() as demo:
+        croper = gr.State()
+        with gr.Row():
+            with gr.Column():
+                edit_prompt = gr.Textbox(lines=1, label="Edit Prompt", value=DEFAULT_EDIT_PROMPT)
+                generate_size = gr.Number(label="Generate Size", value=512)
+            with gr.Column():
+                num_steps = gr.Slider(minimum=1, maximum=100, value=15, step=1, label="Num Steps")
+                guidance_scale = gr.Slider(minimum=0, maximum=30, value=5, step=0.5, label="Guidance Scale")
+            with gr.Column():
+                with gr.Accordion("Advanced Options", open=False):
+                    inpaint_scale = gr.Slider(minimum=0, maximum=3, value=1, step=0.1, label="Inpaint Scale")
+                    canny_scale = gr.Slider(minimum=0, maximum=3, value=0.8, step=0.1, label="Canny Scale")
+                    mask_expansion = gr.Number(label="Mask Expansion", value=50, visible=True)
+                    mask_dilation = gr.Slider(minimum=0, maximum=10, value=2, step=1, label="Mask Dilation")
+                    seed = gr.Number(label="Seed", value=8)
+                    category = gr.Textbox(label="Category", value=DEFAULT_CATEGORY, visible=False)
+                g_btn = gr.Button("Edit Image")
+                
+        with gr.Row():
+            with gr.Column():
+                input_image = gr.Image(label="Input Image", type="pil")
+            with gr.Column():
+                restored_image = gr.Image(label="Restored Image", type="pil", interactive=False)
+            with gr.Column():
+                origin_area_image = gr.Image(label="Origin Area Image", type="pil", interactive=False)
+                generated_image = gr.Image(label="Generated Image", type="pil", interactive=False)
+                generated_cost = gr.Textbox(label="Time cost by step (ms):", visible=True, interactive=False)
+                mask_image = gr.Image(label="Mask Image", type="pil", interactive=False)
+        
+        g_btn.click(
+            fn=segment_image_withmask,
+            inputs=[input_image, category, generate_size, mask_expansion, mask_dilation],
+            outputs=[origin_area_image, mask_image, croper],
+        ).success(
+            fn=image_to_image,
+            inputs=[origin_area_image, mask_image, edit_prompt,seed, num_steps, guidance_scale, generate_size, inpaint_scale, canny_scale],
+            outputs=[generated_image, generated_cost],
+        ).success(
+            fn=restore_result,
+            inputs=[croper, category, generated_image],
+            outputs=[restored_image],
+        )
+
+    return demo

app_haircolor_inpaint_15.py

@@ -17,6 +17,9 @@ from diffusers import (
 
 from controlnet_aux import (
     CannyDetector,
+    LineartDetector,
+    PidiNetDetector,
+    HEDdetector,
 )
 
 BASE_MODEL = "stable-diffusion-v1-5/stable-diffusion-v1-5"
@@ -29,14 +32,22 @@ DEFAULT_NEGATIVE_PROMPT = "worst quality, normal quality, low quality, low res,
 DEFAULT_CATEGORY = "hair"
 
 canny_detector = CannyDetector()
+lineart_detector = LineartDetector.from_pretrained("lllyasviel/Annotators")
+lineart_detector = lineart_detector.to(DEVICE)
+
+pidiNet_detector = PidiNetDetector.from_pretrained('lllyasviel/Annotators')
+pidiNet_detector = pidiNet_detector.to(DEVICE)
+
+hed_detector = HEDdetector.from_pretrained('lllyasviel/Annotators')
+hed_detector = hed_detector.to(DEVICE)
 
 controlnet = [
     ControlNetModel.from_pretrained(
-        "lllyasviel/control_v11p_sd15_inpaint", 
+        "lllyasviel/control_v11p_sd15_lineart", 
         torch_dtype=torch.float16,
     ),
     ControlNetModel.from_pretrained(
-        "lllyasviel/control_v11p_sd15_canny",
+        "lllyasviel/control_v11p_sd15_softedge",
         torch_dtype=torch.float16,
     ),
 ]
@@ -62,15 +73,17 @@ def image_to_image(
     num_steps: int,
     guidance_scale: float,
     generate_size: int,
-    inpaint_scale: float = 1.0,
-    canny_scale: float = 0.5,
+    cond_scale1: float = 1.2,
+    cond_scale2: float = 1.2,
 ):
     run_task_time = 0
     time_cost_str = ''
     run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
-    canny_image = canny_detector(input_image, int(generate_size*0.375), generate_size)
-    inpaint_image = make_inpaint_condition(input_image, mask_image)
-    control_image = [inpaint_image, canny_image]
+    # canny_image = canny_detector(input_image, int(generate_size*0.375), generate_size)
+    lineart_image = lineart_detector(input_image, int(generate_size*0.375), generate_size)
+    run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
+    pidiNet_image = pidiNet_detector(input_image, int(generate_size*0.5), generate_size)
+    control_image = [lineart_image, pidiNet_image]
 
     generator = torch.Generator(device=DEVICE).manual_seed(seed)
     generated_image = basepipeline(
@@ -84,7 +97,7 @@ def image_to_image(
         width=generate_size,
         guidance_scale=guidance_scale,
         num_inference_steps=num_steps,
-        controlnet_conditioning_scale=[inpaint_scale, canny_scale]
+        controlnet_conditioning_scale=[cond_scale1, cond_scale2]
     ).images[0]
     
     run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
@@ -124,8 +137,8 @@ def create_demo() -> gr.Blocks:
                 guidance_scale = gr.Slider(minimum=0, maximum=30, value=5, step=0.5, label="Guidance Scale")
             with gr.Column():
                 with gr.Accordion("Advanced Options", open=False):
-                    inpaint_scale = gr.Slider(minimum=0, maximum=3, value=1, step=0.1, label="Inpaint Scale")
-                    canny_scale = gr.Slider(minimum=0, maximum=3, value=0.8, step=0.1, label="Canny Scale")
+                    cond_scale1 = gr.Slider(minimum=0, maximum=3, value=1.2, step=0.1, label="Lineart Scale")
+                    cond_scale2 = gr.Slider(minimum=0, maximum=3, value=1.2, step=0.1, label="PidiNet Scale")
                     mask_expansion = gr.Number(label="Mask Expansion", value=50, visible=True)
                     mask_dilation = gr.Slider(minimum=0, maximum=10, value=2, step=1, label="Mask Dilation")
                     seed = gr.Number(label="Seed", value=8)
@@ -149,7 +162,7 @@ def create_demo() -> gr.Blocks:
             outputs=[origin_area_image, mask_image, croper],
         ).success(
             fn=image_to_image,
-            inputs=[origin_area_image, mask_image, edit_prompt,seed, num_steps, guidance_scale, generate_size, inpaint_scale, canny_scale],
+            inputs=[origin_area_image, mask_image, edit_prompt,seed, num_steps, guidance_scale, generate_size, cond_scale1, cond_scale2],
             outputs=[generated_image, generated_cost],
         ).success(
             fn=restore_result,