Update app.py

app.py

@@ -7,16 +7,46 @@ from PIL import Image
 import io
 import tempfile
 from pathlib import Path
+import subprocess
+import shutil
 
 # Add notebook directory to path for inference code
 NOTEBOOK_PATH = "notebook"
-if os.path.exists(NOTEBOOK_PATH):
-    sys.path.append(NOTEBOOK_PATH)
+REPO_URL = "https://github.com/facebookresearch/sam-3d-objects"
+
+def ensure_repository():
+    """Ensure the SAM 3D objects repository is cloned"""
+    repo_dir = "sam-3d-objects"
+    
+    if not os.path.exists(repo_dir):
+        print(f"Cloning repository: {REPO_URL}")
+        try:
+            subprocess.run(["git", "clone", REPO_URL, repo_dir], check=True)
+            print("Repository cloned successfully")
+        except subprocess.CalledProcessError as e:
+            print(f"Failed to clone repository: {e}")
+            return False
+        except FileNotFoundError:
+            print("Git not found. Please install git to clone the repository")
+            return False
+    
+    # Add to path
+    if os.path.exists(repo_dir):
+        sys.path.append(repo_dir)
+        if os.path.exists(NOTEBOOK_PATH):
+            sys.path.append(NOTEBOOK_PATH)
+    
+    return True
+
+# Ensure repository is available
+if not ensure_repository():
+    print("Warning: Could not clone repository. Running in limited mode.")
 
 # Import inference code with error handling
 try:
     from inference import Inference, load_image, load_single_mask
     INFERENCE_AVAILABLE = True
+    print("Inference module loaded successfully")
 except ImportError as e:
     print(f"Warning: Could not import inference module: {e}")
     print("Running in demo mode with mock functionality")
@@ -66,13 +96,32 @@ def process_image_to_3d(image, mask=None, seed=42, model_tag="hf"):
             demo_content = create_demo_3d_output()
             return {
                 "status": "demo",
-                "message": "Demo mode - inference module not available",
+                "message": "Demo mode - inference module not available. This is a sample 3D model file.",
                 "file_content": demo_content,
                 "filename": "demo_splat.ply"
             }
         
-        # Initialize inference if not already done
-        config_path = f"checkpoints/{model_tag}/pipeline.yaml"
+        # Check if model directory exists
+        model_dir = f"checkpoints/{model_tag}"
+        if not os.path.exists(model_dir):
+            return {
+                "status": "error",
+                "message": f"Model checkpoint not found at {model_dir}. Please ensure the model is downloaded.",
+                "file_content": None,
+                "filename": None
+            }
+        
+        config_path = f"{model_dir}/pipeline.yaml"
+        if not os.path.exists(config_path):
+            config_path = f"{model_dir}/config.yaml"  # Try alternative config name
+        
+        if not os.path.exists(config_path):
+            return {
+                "status": "error",
+                "message": f"Pipeline configuration not found. Expected at {config_path}",
+                "file_content": None,
+                "filename": None
+            }
         
         # Create temporary files for the uploaded image and mask
         with tempfile.NamedTemporaryFile(suffix=".png", delete=False) as img_temp:
@@ -183,6 +232,18 @@ def create_interface():
         color: #721c24;
         border: 1px solid #f5c6cb;
     }
+    .upload-area {
+        border: 2px dashed #4CAF50 !important;
+        border-radius: 10px !important;
+        padding: 10px !important;
+    }
+    .mask-status {
+        background-color: #e3f2fd;
+        border: 1px solid #2196F3;
+        padding: 5px;
+        border-radius: 5px;
+        font-weight: bold;
+    }
     """
     
     with gr.Blocks(css=css, title="Image to 3D Converter") as demo:
@@ -211,16 +272,25 @@ def create_interface():
                     label="Input Image",
                     type="numpy",
                     image_mode="RGB",
-                    elem_classes=["upload-area"]
+                    elem_classes=["upload-area"],
+                    info="Upload a clear, well-lit image for best results"
                 )
                 
+                gr.HTML("""
+                <div class="upload-section" style="margin-top: 15px;">
+                    <h3>🎭 Optional Mask</h3>
+                    <p>Upload a mask to focus on specific areas</p>
+                </div>
+                """)
+                
                 with gr.Row():
                     mask_upload = gr.Image(
-                        label="Optional Mask",
+                        label="Segmentation Mask",
                         type="numpy", 
                         image_mode="L",
                         image_edit=True,
-                        elem_classes=["upload-area"]
+                        elem_classes=["upload-area"],
+                        info="Optional: Upload a binary mask to segment the object"
                     )
                 
                 mask_status = gr.Textbox(
@@ -232,8 +302,9 @@ def create_interface():
             
             with gr.Column(scale=1):
                 gr.HTML("""
-                <div style="background: #f0f8ff; padding: 20px; border-radius: 10px; margin-bottom: 20px;">
+                <div style="background: #f0f8ff; padding: 20px; border-radius: 10px; margin-bottom: 20px; border: 1px solid #2196F3;">
                     <h3>⚙️ Configuration</h3>
+                    <p>Fine-tune the 3D generation parameters</p>
                 </div>
                 """)
                 
@@ -244,16 +315,23 @@ def create_interface():
                         value=42,
                         step=1,
                         label="Random Seed",
-                        info="Controls the randomness in generation"
+                        info="Controls the randomness in generation. Use different seeds for variations."
                     )
                 
                 model_tag = gr.Dropdown(
                     choices=["hf"],
                     value="hf",
-                    label="Model Tag",
-                    info="Select the model configuration"
+                    label="Model Configuration",
+                    info="Select the model configuration for 3D generation"
                 )
                 
+                gr.HTML("""
+                <div style="margin-top: 20px; text-align: center;">
+                    <p><strong>Generation Status:</strong></p>
+                    <p style="color: #666; font-size: 0.9em;">This process may take several minutes depending on image complexity</p>
+                </div>
+                """)
+                
                 run_button = gr.Button(
                     "🚀 Generate 3D Model",
                     variant="primary",
@@ -263,19 +341,21 @@ def create_interface():
         with gr.Row():
             with gr.Column():
                 status_output = gr.Textbox(
-                    label="Status",
+                    label="Generation Status",
                     max_lines=5,
                     interactive=False,
-                    elem_classes=["status-message"]
+                    elem_classes=["status-message"],
+                    info="Real-time updates on the 3D generation process"
                 )
         
         with gr.Row():
             with gr.Column():
                 output_file = gr.File(
-                    label="Download 3D Model",
+                    label="📥 Download 3D Model",
                     file_types=[".ply"],
                     visible=False,
-                    elem_classes=["download-section"]
+                    elem_classes=["download-section"],
+                    info="Your generated 3D model will appear here for download"
                 )
         
         # Wire up the interface
@@ -291,24 +371,46 @@ def create_interface():
             outputs=[status_output, output_file, gr.File()]
         )
         
-        # Examples section
+        # Examples and instructions section
         gr.HTML("""
-        <div style="margin-top: 40px; text-align: center;">
+        <div style="margin-top: 40px; text-align: center; background: #f8f9fa; padding: 30px; border-radius: 15px;">
             <h3>📖 How to Use</h3>
-            <div style="display: flex; justify-content: space-around; margin-top: 20px; flex-wrap: wrap;">
-                <div style="max-width: 300px; padding: 20px; background: #f8f9fa; border-radius: 10px; margin: 10px;">
+            <div style="display: flex; justify-content: space-around; margin-top: 20px; flex-wrap: wrap; gap: 20px;">
+                <div style="max-width: 280px; padding: 20px; background: white; border-radius: 10px; margin: 10px; box-shadow: 0 2px 10px rgba(0,0,0,0.1);">
                     <h4>1. Upload Image</h4>
-                    <p>Choose a clear, well-lit image for best results</p>
+                    <p>Choose a clear, well-lit image for best results. The image should show the object you want to convert to 3D.</p>
                 </div>
-                <div style="max-width: 300px; padding: 20px; background: #f8f9fa; border-radius: 10px; margin: 10px;">
+                <div style="max-width: 280px; padding: 20px; background: white; border-radius: 10px; margin: 10px; box-shadow: 0 2px 10px rgba(0,0,0,0.1);">
                     <h4>2. Add Mask (Optional)</h4>
-                    <p>Upload a mask to focus on specific areas</p>
+                    <p>Upload a mask to focus on specific areas. This helps the model understand what part of the image to convert.</p>
                 </div>
-                <div style="max-width: 300px; padding: 20px; background: #f8f9fa; border-radius: 10px; margin: 10px;">
+                <div style="max-width: 280px; padding: 20px; background: white; border-radius: 10px; margin: 10px; box-shadow: 0 2px 10px rgba(0,0,0,0.1);">
                     <h4>3. Generate</h4>
-                    <p>Click generate and wait for your 3D model</p>
+                    <p>Click generate and wait for your 3D model. The process typically takes 1-5 minutes.</p>
                 </div>
             </div>
+            
+            <div style="margin-top: 30px; padding: 20px; background: white; border-radius: 10px; text-align: left; max-width: 800px; margin-left: auto; margin-right: auto;">
+                <h4>💡 Tips for Better Results:</h4>
+                <ul style="text-align: left;">
+                    <li>Use high-quality, well-lit images</li>
+                    <li>Ensure the object is clearly visible and not occluded</li>
+                    <li>Use masks to isolate specific objects in complex scenes</li>
+                    <li>Try different random seeds for variations</li>
+                    <li>Complex objects may take longer to process</li>
+                </ul>
+            </div>
+        </div>
+        """)
+        
+        # System information
+        gr.HTML(f"""
+        <div style="margin-top: 30px; padding: 15px; background: #e8f5e8; border-radius: 10px; text-align: center;">
+            <p><strong>System Status:</strong></p>
+            <p style="font-size: 0.9em; color: #666;">
+                Inference Module: {"✓ Available" if INFERENCE_AVAILABLE else "✗ Demo Mode"} | 
+                SAM 3D Repository: {"✓ Cloned" if os.path.exists("sam-3d-objects") else "✗ Not Available"}
+            </p>
         </div>
         """)
     
@@ -336,4 +438,4 @@ if __name__ == "__main__":
         show_error=True,
         debug=True,
         inbrowser=True
-    )
+    )