Update app_merged.py

app_merged.py

@@ -58,6 +58,18 @@ from depth_anything_v2.dpt import DepthAnythingV2
 import httpx
 
 
+import gradio as gr
+import torch
+from diffusers import FluxFillPipeline
+from diffusers.utils import load_image
+from PIL import Image, ImageDraw
+import numpy as np
+import spaces
+from huggingface_hub import hf_hub_download
+
+
+
+
 client = httpx.Client(timeout=httpx.Timeout(10.0))  # Set timeout to 10 seconds
 NUM_VIEWS = 6
 HEIGHT = 768
@@ -140,6 +152,13 @@ hf_hub_download(repo_id="black-forest-labs/FLUX.1-dev", filename="ae.safetensors
 hf_hub_download(repo_id="comfyanonymous/flux_text_encoders", filename="clip_l.safetensors", local_dir="models/text_encoders")
 t5_path = hf_hub_download(repo_id="comfyanonymous/flux_text_encoders", filename="t5xxl_fp16.safetensors", local_dir="models/text_encoders/t5")
 
+fill_pipe = FluxFillPipeline.from_pretrained(
+    "black-forest-labs/FLUX.1-Fill-dev",
+    torch_dtype=torch.bfloat16
+).to("cuda")
+
+
+
 
 sd15_name = 'stablediffusionapi/realistic-vision-v51'
 tokenizer = CLIPTokenizer.from_pretrained(sd15_name, subfolder="tokenizer")
@@ -191,6 +210,183 @@ with torch.no_grad():
 unet_original_forward = unet.forward
 
 
+def can_expand(source_width, source_height, target_width, target_height, alignment):
+    if alignment in ("Left", "Right") and source_width >= target_width:
+        return False
+    if alignment in ("Top", "Bottom") and source_height >= target_height:
+        return False
+    return True
+
+def prepare_image_and_mask(image, width, height, overlap_percentage, resize_option, custom_resize_percentage, alignment, overlap_left, overlap_right, overlap_top, overlap_bottom):
+    target_size = (width, height)
+
+    scale_factor = min(target_size[0] / image.width, target_size[1] / image.height)
+    new_width = int(image.width * scale_factor)
+    new_height = int(image.height * scale_factor)
+    
+    source = image.resize((new_width, new_height), Image.LANCZOS)
+
+    if resize_option == "Full":
+        resize_percentage = 100
+    elif resize_option == "75%":
+        resize_percentage = 75
+    elif resize_option == "50%":
+        resize_percentage = 50
+    elif resize_option == "33%":
+        resize_percentage = 33
+    elif resize_option == "25%":
+        resize_percentage = 25
+    else:  # Custom
+        resize_percentage = custom_resize_percentage
+
+    # Calculate new dimensions based on percentage
+    resize_factor = resize_percentage / 100
+    new_width = int(source.width * resize_factor)
+    new_height = int(source.height * resize_factor)
+
+    # Ensure minimum size of 64 pixels
+    new_width = max(new_width, 64)
+    new_height = max(new_height, 64)
+
+    # Resize the image
+    source = source.resize((new_width, new_height), Image.LANCZOS)
+
+    # Calculate the overlap in pixels based on the percentage
+    overlap_x = int(new_width * (overlap_percentage / 100))
+    overlap_y = int(new_height * (overlap_percentage / 100))
+
+    # Ensure minimum overlap of 1 pixel
+    overlap_x = max(overlap_x, 1)
+    overlap_y = max(overlap_y, 1)
+
+    # Calculate margins based on alignment
+    if alignment == "Middle":
+        margin_x = (target_size[0] - new_width) // 2
+        margin_y = (target_size[1] - new_height) // 2
+    elif alignment == "Left":
+        margin_x = 0
+        margin_y = (target_size[1] - new_height) // 2
+    elif alignment == "Right":
+        margin_x = target_size[0] - new_width
+        margin_y = (target_size[1] - new_height) // 2
+    elif alignment == "Top":
+        margin_x = (target_size[0] - new_width) // 2
+        margin_y = 0
+    elif alignment == "Bottom":
+        margin_x = (target_size[0] - new_width) // 2
+        margin_y = target_size[1] - new_height
+
+    # Adjust margins to eliminate gaps
+    margin_x = max(0, min(margin_x, target_size[0] - new_width))
+    margin_y = max(0, min(margin_y, target_size[1] - new_height))
+
+    # Create a new background image and paste the resized source image
+    background = Image.new('RGB', target_size, (255, 255, 255))
+    background.paste(source, (margin_x, margin_y))
+
+    # Create the mask
+    mask = Image.new('L', target_size, 255)
+    mask_draw = ImageDraw.Draw(mask)
+
+    # Calculate overlap areas
+    white_gaps_patch = 2
+
+    left_overlap = margin_x + overlap_x if overlap_left else margin_x + white_gaps_patch
+    right_overlap = margin_x + new_width - overlap_x if overlap_right else margin_x + new_width - white_gaps_patch
+    top_overlap = margin_y + overlap_y if overlap_top else margin_y + white_gaps_patch
+    bottom_overlap = margin_y + new_height - overlap_y if overlap_bottom else margin_y + new_height - white_gaps_patch
+    
+    if alignment == "Left":
+        left_overlap = margin_x + overlap_x if overlap_left else margin_x
+    elif alignment == "Right":
+        right_overlap = margin_x + new_width - overlap_x if overlap_right else margin_x + new_width
+    elif alignment == "Top":
+        top_overlap = margin_y + overlap_y if overlap_top else margin_y
+    elif alignment == "Bottom":
+        bottom_overlap = margin_y + new_height - overlap_y if overlap_bottom else margin_y + new_height
+
+    # Draw the mask
+    mask_draw.rectangle([
+        (left_overlap, top_overlap),
+        (right_overlap, bottom_overlap)
+    ], fill=0)
+
+    return background, mask
+
+@spaces.GPU
+def inpaint(image, width, height, overlap_percentage, num_inference_steps, resize_option, custom_resize_percentage, prompt_input, alignment, overlap_left, overlap_right, overlap_top, overlap_bottom, progress=gr.Progress(track_tqdm=True)):
+    
+    background, mask = prepare_image_and_mask(image, width, height, overlap_percentage, resize_option, custom_resize_percentage, alignment, overlap_left, overlap_right, overlap_top, overlap_bottom)
+    
+    if not can_expand(background.width, background.height, width, height, alignment):
+        alignment = "Middle"
+
+    cnet_image = background.copy()
+    cnet_image.paste(0, (0, 0), mask)
+
+    final_prompt = prompt_input
+
+    #generator = torch.Generator(device="cuda").manual_seed(42)
+
+    fill_result = fill_pipe(
+        prompt=final_prompt,
+        height=height,
+        width=width,
+        image=cnet_image,
+        mask_image=mask,
+        num_inference_steps=num_inference_steps,
+        guidance_scale=30,
+    ).images[0]
+   
+    fill_result = result.convert("RGBA")
+    cnet_image.paste(result, (0, 0), mask)
+
+    return cnet_image, background
+
+def preview_image_and_mask(image, width, height, overlap_percentage, resize_option, custom_resize_percentage, alignment, overlap_left, overlap_right, overlap_top, overlap_bottom):
+    background, mask = prepare_image_and_mask(image, width, height, overlap_percentage, resize_option, custom_resize_percentage, alignment, overlap_left, overlap_right, overlap_top, overlap_bottom)
+    
+    preview = background.copy().convert('RGBA')
+    red_overlay = Image.new('RGBA', background.size, (255, 0, 0, 64))
+    red_mask = Image.new('RGBA', background.size, (0, 0, 0, 0))
+    red_mask.paste(red_overlay, (0, 0), mask)
+    preview = Image.alpha_composite(preview, red_mask)
+    
+    return preview
+
+def clear_result():
+    return gr.update(value=None)
+
+def preload_presets(target_ratio, ui_width, ui_height):
+    if target_ratio == "9:16":
+        return 720, 1280, gr.update()
+    elif target_ratio == "16:9":
+        return 1280, 720, gr.update()
+    elif target_ratio == "1:1":
+        return 1024, 1024, gr.update()
+    elif target_ratio == "Custom":
+        return ui_width, ui_height, gr.update(open=True)
+
+def select_the_right_preset(user_width, user_height):
+    if user_width == 720 and user_height == 1280:
+        return "9:16"
+    elif user_width == 1280 and user_height == 720:
+        return "16:9"
+    elif user_width == 1024 and user_height == 1024:
+        return "1:1"
+    else:
+        return "Custom"
+
+def toggle_custom_resize_slider(resize_option):
+    return gr.update(visible=(resize_option == "Custom"))
+
+def update_history(new_image, history):
+    if history is None:
+        history = []
+    history.insert(0, new_image)
+    return history
+
+
 def enable_efficient_attention():
     if XFORMERS_AVAILABLE:
         try:
@@ -1390,6 +1586,77 @@ with gr.Blocks() as app:
                 
             with gr.Column():
                 result_gallery = gr.Gallery(height=832, object_fit='contain', label='Outputs')
+                with gr.Group():
+                    gr.Markdown("Outpaint")
+                    with gr.Row():
+                        with gr.Column(scale=2):
+                            prompt_fill = gr.Textbox(label="Prompt (Optional)")
+                        with gr.Column(scale=1):
+                            fill_button = gr.Button("Generate")
+                    target_ratio = gr.Radio(
+                        label="Image Ratio",
+                        choices=["9:16", "16:9", "1:1", "Custom"],
+                        value="9:16",
+                        scale=3
+                    )
+                    alignment_dropdown = gr.Dropdown(
+                        choices=["Middle", "Left", "Right", "Top", "Bottom"],
+                        value="Middle",
+                        label="Alignment",
+                    )
+                    resize_option = gr.Radio(
+                        label="Resize input image",
+                        choices=["Full", "75%", "50%", "33%", "25%", "Custom"],
+                        value="75%"
+                    )
+                    custom_resize_percentage = gr.Slider(
+                        label="Custom resize (%)",
+                        minimum=1,
+                        maximum=100,
+                        step=1,
+                        value=50,
+                        visible=False
+                    )
+
+                    fill_result = gr.Image(
+                        interactive=False,
+                        label="Generated Image",
+                    )
+                    
+                    with gr.Accordion(label="Advanced settings", open=False) as settings_panel:
+                        with gr.Column():
+                            with gr.Row():
+                                width_slider = gr.Slider(
+                                    label="Target Width",
+                                    minimum=720,
+                                    maximum=1536,
+                                    step=8,
+                                    value=720,
+                                )
+                                height_slider = gr.Slider(
+                                    label="Target Height",
+                                    minimum=720,
+                                    maximum=1536,
+                                    step=8,
+                                    value=1280,
+                                )
+                            
+                            num_inference_steps = gr.Slider(label="Steps", minimum=2, maximum=50, step=1, value=28)
+                            with gr.Group():
+                                overlap_percentage = gr.Slider(
+                                    label="Mask overlap (%)",
+                                    minimum=1,
+                                    maximum=50,
+                                    value=10,
+                                    step=1
+                                )
+                                with gr.Row():
+                                    overlap_top = gr.Checkbox(label="Overlap Top", value=True)
+                                    overlap_right = gr.Checkbox(label="Overlap Right", value=True)
+                                with gr.Row():
+                                    overlap_left = gr.Checkbox(label="Overlap Left", value=True)
+                                    overlap_bottom = gr.Checkbox(label="Overlap Bottom", value=True)
+                            
 
                 with gr.Row():
                     with gr.Group():
@@ -1451,6 +1718,70 @@ with gr.Blocks() as app:
         example_quick_prompts.click(lambda x, y: ', '.join(y.split(', ')[:2] + [x[0]]), inputs=[example_quick_prompts, prompt], outputs=prompt, show_progress=False, queue=False)
         example_quick_subjects.click(lambda x: x[0], inputs=example_quick_subjects, outputs=prompt, show_progress=False, queue=False)
 
+        # def use_output_as_input(output_image):
+        # return output_image
+
+        # use_as_input_button.click(
+        #     fn=use_output_as_input,
+        #     inputs=[fill_result],
+        #     outputs=[input_image]
+        # )
+        
+        target_ratio.change(
+            fn=preload_presets,
+            inputs=[target_ratio, width_slider, height_slider],
+            outputs=[width_slider, height_slider, settings_panel],
+            queue=False
+        )
+    
+        width_slider.change(
+            fn=select_the_right_preset,
+            inputs=[width_slider, height_slider],
+            outputs=[target_ratio],
+            queue=False
+        )
+    
+        height_slider.change(
+            fn=select_the_right_preset,
+            inputs=[width_slider, height_slider],
+            outputs=[target_ratio],
+            queue=False
+        )
+    
+        resize_option.change(
+            fn=toggle_custom_resize_slider,
+            inputs=[resize_option],
+            outputs=[custom_resize_percentage],
+            queue=False
+        )
+        
+        fill_button.click(
+            fn=clear_result,
+            inputs=None,
+            outputs=fill_result,
+        ).then(
+            fn=inpaint,
+            inputs=[result_gallery, width_slider, height_slider, overlap_percentage, num_inference_steps,
+                    resize_option, custom_resize_percentage, prompt_fill, alignment_dropdown,
+                    overlap_left, overlap_right, overlap_top, overlap_bottom],
+            outputs=[fill_result])
+        # ).then(
+        #     fn=lambda: gr.update(visible=True),
+        #     inputs=None,
+        #     outputs=use_as_input_button,
+        # )
+
+        prompt_fill.submit(
+            fn=clear_result,
+            inputs=None,
+            outputs=fill_result,
+        ).then(
+            fn=inpaint,
+            inputs=[result_gallery, width_slider, height_slider, overlap_percentage, num_inference_steps,
+                    resize_option, custom_resize_percentage, prompt_fill, alignment_dropdown,
+                    overlap_left, overlap_right, overlap_top, overlap_bottom],
+            outputs=[fill_result])
+
 
         def convert_to_pil(image):
             try: