Update app.py

app.py

@@ -9,7 +9,6 @@ import os
 
 # --- 1. SETTINGS & MODEL ---
 DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
-# Using Depth Anything V2 for maximum compatibility
 CHECKPOINT = "depth-anything/Depth-Anything-V2-Small-hf"
 
 processor = AutoImageProcessor.from_pretrained(CHECKPOINT)
@@ -23,7 +22,6 @@ def process_to_3d(input_image):
     inputs = processor(images=input_image, return_tensors="pt").to(DEVICE)
     with torch.no_grad():
         outputs = model(**inputs)
-        # Resize depth map to match original image resolution
         depth = torch.nn.functional.interpolate(
             outputs.predicted_depth.unsqueeze(1),
             size=input_image.size[::-1],
@@ -35,10 +33,10 @@ def process_to_3d(input_image):
     rgb = np.array(input_image)
     x, y = np.meshgrid(np.arange(width), np.arange(height))
     
-    # Scale depth to a standard 3D unit range
+    # Scale depth (Z-axis) for a clean 3D range
     z = (depth / depth.max()) * 10.0 
     
-    # Projection math (pinhole camera model)
+    # Projection math
     focal_length = width 
     x_coords = (x - width / 2) * z / focal_length
     y_coords = (y - height / 2) * z / focal_length
@@ -46,50 +44,47 @@ def process_to_3d(input_image):
     points = np.stack((x_coords, y_coords, z), axis=-1).reshape(-1, 3)
     colors = rgb.reshape(-1, 3) / 255.0
 
-    # --- 4. THE SPLAT TRICK (Open3D) ---
+    # --- 4. CENTERING & VOXELIZATION ---
     pcd = o3d.geometry.PointCloud()
     pcd.points = o3d.utility.Vector3dVector(points)
     pcd.colors = o3d.utility.Vector3dVector(colors)
 
-    # Centering: Move the model so its 3D center is at (0, 0, 0)
-    # This ensures the camera rotates around the object, not the corner.
+    # Centering: Critical for the camera to lock onto the model
     center = pcd.get_center()
     pcd.translate(-center)
 
-    # Voxelization: This merges tiny points into larger "Splats"
-    # Adjust voxel_size to make the model more or less "dense"
-    pcd = pcd.voxel_down_sample(voxel_size=0.05) 
+    # Voxelization: Merges points into larger "splats" for solid visibility
+    pcd = pcd.voxel_down_sample(voxel_size=0.04) 
 
-    # --- 5. EXPORT ---
+    # --- 5. EXPORT AS .PLY ---
     temp_dir = tempfile.gettempdir()
-    # Saving as .ply (Gradio 5+ renders binary PLY as splats in Solid mode)
-    output_path = os.path.join(temp_dir, "model_output.ply")
+    output_path = os.path.join(temp_dir, "model.ply")
+    
+    # write_ascii=False saves it in Binary format (required for fast web loading)
     o3d.io.write_point_cloud(output_path, pcd, write_ascii=False)
     
     return output_path, output_path
 
 # --- 6. GRADIO UI ---
-with gr.Blocks(theme=gr.themes.Soft()) as demo:
-    gr.Markdown("# 🌌 3D Gaussian Splat Generator")
-    gr.Markdown("Transform any 2D image into a centered, solid-looking 3D Splat.")
+with gr.Blocks(theme=gr.themes.Default()) as demo:
+    gr.Markdown("# 🌊 Depth Anything Splat Creator")
     
     with gr.Row():
         with gr.Column(scale=1):
             img_input = gr.Image(type="pil", label="Input Image")
-            run_btn = gr.Button("🔨 Build 3D Splat", variant="primary")
+            run_btn = gr.Button("🔨 Generate .PLY Splat", variant="primary")
         
         with gr.Column(scale=2):
-            # display_mode="solid" tells Gradio to render the points as Gaussians
-            # camera_position=(alpha, beta, radius)
             view_3d = gr.Model3D(
                 label="3D Viewport",
-                display_mode="solid", 
+                display_mode="solid", # Renders PLY points as Gaussians
                 camera_position=(0, 90, 15), 
                 clear_color=(0.0, 0.0, 0.0, 1.0)
             )
-            dl_btn = gr.DownloadButton("💾 Download Model (.PLY)")
+            # Explicitly set the download button
+            dl_btn = gr.DownloadButton("💾 Download .PLY File")
 
-    # Define behavior
+    # Link the logic
     run_btn.click(
         fn=process_to_3d,
         inputs=[img_input],