feat: upgrade to BiRefNet SOTA background removal model

- Switch default model from U2-Net to BiRefNet (state-of-the-art)
- Add 6 model options: BiRefNet, BiRefNet Lite, BiRefNet Portrait,
  BRIA RMBG, IS-Net, U2-Net
- Add animated progress bar with stage indicators (loading model,
  analyzing, removing, refining edges) and elapsed timer
- Add intuitive background color presets (Transparent, White, Black,
  Red, Green, Blue) as clickable buttons + custom color picker
- Handle background color compositing in Python (PIL alpha composite)
- Add checkerboard pattern to before/after slider for transparency
- Pre-bake BiRefNet model (973MB) in Docker image for instant use

packages/ai/python/remove_bg.py

@@ -1,4 +1,4 @@
-"""Background removal using rembg."""
+"""Background removal using rembg with state-of-the-art BiRefNet models."""
 import sys
 import json
 
@@ -8,25 +8,48 @@ def main():
     output_path = sys.argv[2]
     settings = json.loads(sys.argv[3]) if len(sys.argv) > 3 else {}
 
-    model = settings.get("model", "u2net")
+    model = settings.get("model", "birefnet-general")
+    bg_color = settings.get("backgroundColor", "")
 
     try:
-        from rembg import remove
+        from rembg import remove, new_session
+        from PIL import Image
+        import io
+
+        print(json.dumps({"progress": "loading_model"}), flush=True)
+
+        # Create a session with the selected model
+        session = new_session(model)
 
         with open(input_path, "rb") as f:
             input_data = f.read()
 
-        # Try with alpha matting first for better edges, but fall back
-        # without it if the image triggers the known rembg matting error
+        print(json.dumps({"progress": "processing"}), flush=True)
+
+        # Try with alpha matting first for better edges
         try:
             output_data = remove(
                 input_data,
+                session=session,
                 alpha_matting=True,
                 alpha_matting_foreground_threshold=240,
                 alpha_matting_background_threshold=10,
             )
         except Exception:
-            output_data = remove(input_data)
+            output_data = remove(input_data, session=session)
+
+        # If a background color is specified, composite onto it
+        if bg_color and bg_color.startswith("#"):
+            img = Image.open(io.BytesIO(output_data)).convert("RGBA")
+            hex_color = bg_color.lstrip("#")
+            r = int(hex_color[0:2], 16)
+            g = int(hex_color[2:4], 16)
+            b = int(hex_color[4:6], 16)
+            bg = Image.new("RGBA", img.size, (r, g, b, 255))
+            bg.paste(img, mask=img.split()[3])
+            buf = io.BytesIO()
+            bg.save(buf, format="PNG")
+            output_data = buf.getvalue()
 
         with open(output_path, "wb") as f:
             f.write(output_data)