prompt in image2image

app.py

@@ -40,7 +40,10 @@ def text_to_image(prompt, negative_prompt, guidance_scale, num_inference_steps):
 
 
 # Function for Image-to-Image using Dynamic UNet
-def apply_dynamic_unet(init_image, strength):
+def apply_dynamic_unet(init_image, prompt, strength):
+    # Placeholder for prompt-based logic
+    print(f"Received prompt: {prompt}")  # You can define prompt-based rules here.
+
     with torch.no_grad():
         image_tensor = preprocess(init_image).unsqueeze(0).to(device)
         output = unet_model(image_tensor)["out"][0]
@@ -73,9 +76,16 @@ with gr.Blocks(theme='Respair/[email protected]') as demo:
             outputs=text_output,
         )
 
+    # Gradio Tab with Prompt
     with gr.Tab("Image-to-Image"):
+        gr.Markdown(
+            "**Transform uploaded images using a dynamic UNet model.**\n"
+            "Provide a prompt to describe the transformation and use the `Blend Strength` slider to adjust blending."
+        )
         with gr.Row():
             init_image = gr.Image(type="pil", label="Upload Initial Image")
+        with gr.Row():
+            img_prompt = gr.Textbox(label="Prompt", placeholder="Describe the transformation (optional)...")
         with gr.Row():
             strength = gr.Slider(0.1, 1.0, value=0.75, step=0.05, label="Blend Strength")
         with gr.Row():
@@ -83,6 +93,6 @@ with gr.Blocks(theme='Respair/[email protected]') as demo:
         with gr.Row():
             img_output = gr.Image(label="Modified Image")
 
-        img_generate_btn.click(apply_dynamic_unet, inputs=[init_image, strength], outputs=img_output)
+        img_generate_btn.click(apply_dynamic_unet, inputs=[init_image, img_prompt, strength], outputs=img_output)
 
 demo.launch(share=True)