Update app.py

app.py

@@ -1,16 +1,18 @@
-from transformers import CLIPSegProcessor, CLIPSegForImageSegmentation
-import gradio as gr
+from flask import Flask, request, jsonify, render_template
 from PIL import Image
+import base64
+from io import BytesIO
+from transformers import CLIPSegProcessor, CLIPSegForImageSegmentation
 import torch
+import numpy as np
 import matplotlib.pyplot as plt
 import cv2
-import torch
-import numpy as np
+
+app = Flask(__name__)
 
 processor = CLIPSegProcessor.from_pretrained("CIDAS/clipseg-rd64-refined")
 model = CLIPSegForImageSegmentation.from_pretrained("CIDAS/clipseg-rd64-refined")
 
-
 def process_image(image, prompt, threhsold, alpha_value, draw_rectangles):
     inputs = processor(
         text=prompt, images=image, padding="max_length", return_tensors="pt"
@@ -61,62 +63,34 @@ def process_image(image, prompt, threhsold, alpha_value, draw_rectangles):
     output_image.paste(image, mask=bmask)
 
     return fig, mask, output_image
+    
+@app.route('/')
+def index():
+    return "Hello, World! clipseg2"
 
+@app.route('/api/mask_image', methods=['POST'])
+def mask_image_api():
+    data = request.get_json()
 
-title = "Interactive demo: zero-shot image segmentation with CLIPSeg"
-description = "Demo for using CLIPSeg, a CLIP-based model for zero- and one-shot image segmentation. To use it, simply upload an image and add a text to mask (identify in the image), or use one of the examples below and click 'submit'. Results will show up in a few seconds."
-article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2112.10003'>CLIPSeg: Image Segmentation Using Text and Image Prompts</a> | <a href='https://huggingface.co/docs/transformers/main/en/model_doc/clipseg'>HuggingFace docs</a></p>"
+    base64_image = data.get('base64_image', '')
+    prompt = data.get('prompt', '')
+    threshold = data.get('threshold', 0.4)
+    alpha_value = data.get('alpha_value', 0.5)
+    draw_rectangles = data.get('draw_rectangles', False)
 
+    # Decode base64 image
+    image_data = base64.b64decode(base64_image.split(',')[1])
+    image = Image.open(BytesIO(image_data))
 
-with gr.Blocks() as demo:
-    gr.Markdown("# CLIPSeg: Image Segmentation Using Text and Image Prompts")
-    gr.Markdown(article)
-    gr.Markdown(description)
-    gr.Markdown(
-        "*Example images are taken from the [ImageNet-A](https://paperswithcode.com/dataset/imagenet-a) dataset*"
-    )
+    # Process the image
+    _, _, output_image = process_image(image, prompt, threshold, alpha_value, draw_rectangles)
 
-    with gr.Row():
-        with gr.Column():
-            input_image = gr.Image(type="pil")
-            input_prompt = gr.Textbox(label="Please describe what you want to identify")
-            input_slider_T = gr.Slider(
-                minimum=0, maximum=1, value=0.4, label="Threshold"
-            )
-            input_slider_A = gr.Slider(minimum=0, maximum=1, value=0.5, label="Alpha")
-            draw_rectangles = gr.Checkbox(label="Draw rectangles")
-            btn_process = gr.Button(label="Process")
-
-        with gr.Column():
-            output_plot = gr.Plot(label="Segmentation Result")
-            output_mask = gr.Image(label="Mask")
-            output_image = gr.Image(label="Output Image")
-
-    btn_process.click(
-        process_image,
-        inputs=[
-            input_image,
-            input_prompt,
-            input_slider_T,
-            input_slider_A,
-            draw_rectangles,
-        ],
-        outputs=[output_plot, output_mask, output_image],api_name="masking"
-    )
+    # Convert the output image to base64
+    buffered = BytesIO()
+    output_image.save(buffered, format="PNG")
+    result_image = base64.b64encode(buffered.getvalue()).decode('utf-8')
 
-    gr.Examples(
-        [
-            ["0.003473_cliff _ cliff_0.51112.jpg", "dog", 0.5, 0.5, True],
-            ["0.001861_submarine _ submarine_0.9862991.jpg", "beacon", 0.55, 0.4, True],
-            ["0.004658_spatula _ spatula_0.35416836.jpg", "banana", 0.4, 0.5, True],
-        ],
-        inputs=[
-            input_image,
-            input_prompt,
-            input_slider_T,
-            input_slider_A,
-            draw_rectangles,
-        ],
-    )
+    return jsonify({'result_image': result_image})
 
-demo.launch()
+if __name__ == '__main__':
+    app.run(host='0.0.0.0', port=7860, debug=True)