Create app.py

app.py

@@ -0,0 +1,54 @@
+import gradio as gr
+from transformers import pipeline
+from diffusers import StableDiffusionPipeline
+import torch
+import wget
+
+# Define the device to use (either "cuda" for GPU or "cpu" for CPU)
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+# Load the models
+caption_image = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large", device=device)
+sd_pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5").to(device)
+
+translator = pipeline(
+    task="translation",
+    model="facebook/nllb-200-distilled-600M",
+    torch_dtype=torch.bfloat16,
+    device=device
+)
+
+# Function to generate images based on the image's caption
+def generate_image_and_translate(image, num_images=1):
+    caption_en = caption_image(image)[0]['generated_text']
+    caption_ar = translator(caption_en, src_lang="eng_Latn", tgt_lang="arb_Arab")[0]['translation_text']
+
+    generated_images = []
+    for _ in range(num_images):
+        generated_image = sd_pipeline(prompt=caption_en).images[0]
+        generated_images.append(generated_image)
+
+    return generated_images, caption_en, caption_ar
+
+# Set up the Gradio interface
+interface = gr.Interface(
+    fn=generate_image_and_translate,
+    inputs=[
+        gr.Image(type="pil", label="Upload Image"),
+        gr.Slider(minimum=1, maximum=10, label="Number of Images", value=1, step=1)
+    ],
+    outputs=[
+        gr.Gallery(label="Generated Images"),
+        gr.Textbox(label="Generated Caption (English)", interactive=False),
+        gr.Textbox(label="Translated Caption (Arabic)", interactive=False)
+    ],
+    title="Image Generation and Translation",
+    description="Upload an image to generate new images based on its caption and translate the caption into Arabic.",
+    examples=[
+        ["sea.jpg", 3]
+    ]
+)
+
+# Launch the Gradio application
+if __name__ == "__main__":
+    interface.launch()