Update app.py

app.py

@@ -8,91 +8,74 @@ import threading
 processor = CLIPSegProcessor.from_pretrained("CIDAS/clipseg-rd64-refined")
 model = CLIPSegForImageSegmentation.from_pretrained("CIDAS/clipseg-rd64-refined")
 
-# Gradio UI
-with gr.Blocks() as demo:
-    gr.Markdown("# CLIPSeg: Image Segmentation Using Text and Image Prompts")
-
-    # Add your article and description here
-    gr.Markdown("Your article goes here")
-    gr.Markdown("Your description goes here")
-
-    with gr.Row():
-        with gr.Column():
-            input_image = gr.Image(type="pil")
-            positive_prompts = gr.Textbox(
-                label="Please describe what you want to identify (comma separated)"
-            )
-            negative_prompts = gr.Textbox(
-                label="Please describe what you want to ignore (comma separated)"
-            )
-            input_slider_T = gr.Slider(
-                minimum=0, maximum=1, value=0.4, label="Threshold"
-            )
-            btn_process = gr.Button(label="Process")
-
-        with gr.Column():
-            output_image = gr.Image(label="Result")
-            output_mask = gr.Image(label="Mask")
-
-    def process_image(image, prompt):
-        inputs = processor(
-            text=prompt, images=image, padding="max_length", return_tensors="pt"
-        )
-        with torch.no_grad():
-            outputs = model(**inputs)
-            preds = outputs.logits
-
-        pred = torch.sigmoid(preds)
-        mat = pred.cpu().numpy()
-        mask = Image.fromarray(np.uint8(mat * 255), "L")
-        mask = mask.convert("RGB")
-        mask = mask.resize(image.size)
-        mask = np.array(mask)[:, :, 0]
-
-        mask_min = mask.min()
-        mask_max = mask.max()
-        mask = (mask - mask_min) / (mask_max - mask_min)
-
-        return mask
-
-    def get_masks(prompts, img, threshold):
-        prompts = prompts.split(",")
-        masks = []
-        for prompt in prompts:
-            mask = process_image(img, prompt)
-            mask = mask > threshold
-            masks.append(mask)
-
-        return masks
-
-    def extract_image(pos_prompts, neg_prompts, img, threshold):
-        positive_masks = get_masks(pos_prompts, img, 0.5)
-        negative_masks = get_masks(neg_prompts, img, 0.5)
-
-        pos_mask = np.any(np.stack(positive_masks), axis=0)
-        neg_mask = np.any(np.stack(negative_masks), axis=0)
-        final_mask = pos_mask & ~neg_mask
-
-        final_mask = Image.fromarray(final_mask.astype(np.uint8) * 255, "L")
-        output_image = Image.new("RGBA", img.size, (0, 0, 0, 0))
-        output_image.paste(img, mask=final_mask)
-
-        return output_image, final_mask
-
-    btn_process.click(
-        extract_image,
-        inputs=[
-            positive_prompts,
-            negative_prompts,
-            input_image,
-            input_slider_T,
-        ],
-        outputs=[output_image, output_mask],
+# Function to process image and generate mask
+def process_image(image, prompt):
+    inputs = processor(
+        text=prompt, images=image, padding="max_length", return_tensors="pt"
     )
-
-# Gradio API Endpoint
-gr.Interface(
-    [positive_prompts, negative_prompts, input_image, input_slider_T],
-    [output_image, output_mask],
-    live=True,
-).launch()
+    with torch.no_grad():
+        outputs = model(**inputs)
+        preds = outputs.logits
+
+    pred = torch.sigmoid(preds)
+    mat = pred.cpu().numpy()
+    mask = Image.fromarray(np.uint8(mat * 255), "L")
+    mask = mask.convert("RGB")
+    mask = mask.resize(image.size)
+    mask = np.array(mask)[:, :, 0]
+
+    mask_min = mask.min()
+    mask_max = mask.max()
+    mask = (mask - mask_min) / (mask_max - mask_min)
+
+    return mask
+
+# Function to get masks from positive or negative prompts
+def get_masks(prompts, img, threshold):
+    prompts = prompts.split(",")
+    masks = []
+    for prompt in prompts:
+        mask = process_image(img, prompt)
+        mask = mask > threshold
+        masks.append(mask)
+
+    return masks
+
+# Function to extract image using positive and negative prompts
+def extract_image(pos_prompts, neg_prompts, img, threshold):
+    positive_masks = get_masks(pos_prompts, img, 0.5)
+    negative_masks = get_masks(neg_prompts, img, 0.5)
+
+    pos_mask = np.any(np.stack(positive_masks), axis=0)
+    neg_mask = np.any(np.stack(negative_masks), axis=0)
+    final_mask = pos_mask & ~neg_mask
+
+    final_mask = Image.fromarray(final_mask.astype(np.uint8) * 255, "L")
+    output_image = Image.new("RGBA", img.size, (0, 0, 0, 0))
+    output_image.paste(img, mask=final_mask)
+
+    return output_image, final_mask
+
+# Define Gradio interface
+iface = gr.Interface(
+    fn=extract_image,
+    inputs=[
+        gr.Textbox(
+            label="Please describe what you want to identify (comma separated)",
+            key="pos_prompts",
+        ),
+        gr.Textbox(
+            label="Please describe what you want to ignore (comma separated)",
+            key="neg_prompts",
+        ),
+        gr.Image(type="pil", label="Input Image", key="img"),
+        gr.Slider(minimum=0, maximum=1, default=0.4, label="Threshold", key="threshold"),
+    ],
+    outputs=[
+        gr.Image(label="Result", key="output_image"),
+        gr.Image(label="Mask", key="output_mask"),
+    ],
+)
+
+# Launch Gradio API
+iface.launch()