Update app.py

app.py

@@ -34,25 +34,12 @@ pipe = prepare_pipeline(
     dtype=dtype,
 )
 
-# remove bg
-birefnet = AutoModelForImageSegmentation.from_pretrained(
-    "ZhengPeng7/BiRefNet", trust_remote_code=True
-)
-birefnet.to(device)
-transform_image = transforms.Compose(
-    [
-        transforms.Resize((1024, 1024)),
-        transforms.ToTensor(),
-        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),
-    ]
-)
-
 
 @spaces.GPU()
 def infer(
     prompt,
     image,
-    do_rembg=True,
+    do_rembg=False,
     seed=42,
     randomize_seed=False,
     guidance_scale=3.0,
@@ -61,10 +48,6 @@ def infer(
     negative_prompt="watermark, ugly, deformed, noisy, blurry, low contrast",
     progress=gr.Progress(track_tqdm=True),
 ):
-    if do_rembg:
-        remove_bg_fn = lambda x: remove_bg(x, birefnet, transform_image, device)
-    else:
-        remove_bg_fn = None
     if randomize_seed:
         seed = random.randint(0, MAX_SEED)
     images, preprocessed_image = run_pipeline(
@@ -77,12 +60,12 @@ def infer(
         num_inference_steps=num_inference_steps,
         guidance_scale=guidance_scale,
         seed=seed,
-        remove_bg_fn=remove_bg_fn,
+        remove_bg_fn=None,
         reference_conditioning_scale=reference_conditioning_scale,
         negative_prompt=negative_prompt,
         device=device,
     )
-    return images, preprocessed_image, seed
+    return images
 
 
 # examples = [