app.py

import os
# 强制安装依赖
os.system("pip install gradio==4.0.0 transformers==4.30.2 torch==1.13.1 pillow==9.4.0")

import gradio as gr
from transformers import pipeline
from PIL import Image

# 加载模型
image_to_text = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base")

def generate_caption(image):
    result = image_to_text(image)[0]
    return f"图像描述：{result['generated_text']}"

with gr.Blocks(title="图像文本描述工具") as demo:
    gr.Markdown("# 图像文本描述工具")
    image_input = gr.Image(type="pil", label="上传图片")
    text_output = gr.Textbox(label="生成描述")
    gr.Button("生成").click(fn=generate_caption, inputs=image_input, outputs=text_output)

if __name__ == "__main__":
    demo.launch()