Update app.py

app.py

@@ -1,27 +1,95 @@
+import os
+import sys
 import gradio as gr
+import subprocess
 import torch
-from diffusers import StableDiffusionPipeline
-
-# Load lightweight SD (works on free tier)
-pipe = StableDiffusionPipeline.from_pretrained(
-    "runwayml/stable-diffusion-v1-5",
-    torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
-    safety_checker=None
-)
-pipe = pipe.to("cuda" if torch.cuda.is_available() else "cpu")
-
-def generate(prompt):
-    image = pipe(prompt, num_inference_steps=20, width=512, height=512).images[0]
-    return image
-
-# Gradio Interface
-iface = gr.Interface(
-    fn=generate,
-    inputs=gr.Textbox(label="Prompt", placeholder="Astronaut in jungle..."),
-    outputs=gr.Image(label="Generated Image"),
-    title="🪄 MagicDrive-Style Demo",
-    description="Street view generation demo"
-)
-
-if __name__ == "__main__":
-    iface.launch()
+from pathlib import Path
+
+# Add MagicDrive as submodule (auto-clones on first run)
+MAGICDRIVE_DIR = Path("./MagicDrive")
+if not MAGICDRIVE_DIR.exists():
+    subprocess.run(["git", "clone", "--recursive", "https://github.com/cure-lab/MagicDrive.git"])
+    sys.path.append(str(MAGICDRIVE_DIR))
+    os.chdir(MAGICDRIVE_DIR)
+
+# Install deps on first run
+subprocess.run(["pip", "install", "-r", "requirements/gui.txt"], capture_output=True)
+subprocess.run(["pip", "install", "gradio", "omegaconf", "hydra-core"], capture_output=True)
+
+# Import MagicDrive modules
+sys.path.insert(0, str(MAGICDRIVE_DIR))
+from demo.interactive_gui import load_model_from, run_pipe
+import glob
+
+# Global model cache
+model_cache = None
+
+def load_magicdrive_model():
+    global model_cache
+    if model_cache is None:
+        # Download minimal pretrained if needed
+        os.makedirs("pretrained/stable-diffusion-v1-5", exist_ok=True)
+        model_dir = "pretrained/SDv1.5mv-rawbox_2023-09-07_18-39_224x400"
+        os.makedirs(model_dir, exist_ok=True)
+        
+        model_cache = load_model_from(model_dir, device="cuda" if torch.cuda.is_available() else "cpu")
+    return model_cache
+
+def generate_image(image_input, prompt, seed=42):
+    if image_input is None:
+        return None, "Upload an image first"
+    
+    model = load_magicdrive_model()
+    cfg, pipe = model
+    
+    # Run MagicDrive inference
+    import numpy as np
+    from PIL import Image
+    import random
+    
+    random.seed(seed)
+    torch.manual_seed(seed)
+    
+    # Convert Gradio image to numpy
+    img_array = np.array(image_input)
+    
+    result = run_pipe(cfg, pipe, img_array, seed)
+    
+    # Convert back to PIL
+    result_img = Image.fromarray((result * 255).astype(np.uint8))
+    return result_img, f"Generated with seed: {seed}"
+
+# Gradio Tabs Interface
+with gr.Blocks(title="🪄 MagicDrive Demo") as demo:
+    gr.Markdown("# 🪄 MagicDrive: Street View Generation")
+    gr.Markdown("Upload image → Generate street view with 3D geometry control")
+    
+    with gr.Tabs():
+        with gr.TabItem("🚗 Street View Generator"):
+            with gr.Row():
+                with gr.Column(scale=1):
+                    input_image = gr.Image(type="pil", label="Input Image")
+                    prompt = gr.Textbox("Street view, realistic, detailed", label="Prompt")
+                    seed = gr.Slider(0, 10000, value=42, label="Seed")
+                    generate_btn = gr.Button("Generate Street View", variant="primary")
+                
+                with gr.Column(scale=1):
+                    output_image = gr.Image(label="Generated Street View")
+                    status = gr.Textbox(label="Status", interactive=False)
+            
+            generate_btn.click(
+                fn=generate_image,
+                inputs=[input_image, prompt, seed],
+                outputs=[output_image, status]
+            )
+        
+        with gr.TabItem("📊 Model Info"):
+            gr.Markdown("""
+            ## MagicDrive Features
+            - Street view synthesis with 3D geometry control
+            - Multi-view consistency
+            - Real-time interactive GUI
+            - Based on: [MagicDrive Paper](https://arxiv.org/abs/2310.10687)
+            """)
+
+demo.launch()