Update app.py

app.py

@@ -3,6 +3,8 @@ import streamlit as st
 from transformers import pipeline
 import math
 import time  # 新增time模块
+from PIL import Image
+import io
 
 # function part
 def split_story_with_delay(story_text, total_duration, num_chunks=5):
@@ -30,12 +32,29 @@ def split_story_with_delay(story_text, total_duration, num_chunks=5):
     chunk_duration = total_duration / len(chunks)
     return list(zip(chunks, [chunk_duration]*len(chunks)))
 
-def img2text(url):
-    # 添加进度提示
-    with st.spinner("🖼️ Analyzing the magic picture..."):
-        image_to_text_model = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base")
-        text = image_to_text_model(url)[0]["generated_text"]
-    return text
+def img2text(uploaded_file):
+    """处理上传文件对象"""
+    try:
+        # 将上传文件转换为PIL Image
+        image = Image.open(io.BytesIO(uploaded_file.getvalue()))
+        
+        # 显示调试信息（可选）
+        st.write(f"✅ 成功读取图片 | 格式: {image.format} | 尺寸: {image.size}")
+        
+        # 创建图片转文本模型
+        image_to_text_model = pipeline(
+            "image-to-text",
+            model="Salesforce/blip-image-captioning-base",
+            device=0 if torch.cuda.is_available() else -1  # 添加GPU支持
+        )
+        
+        # 直接传入PIL Image
+        result = image_to_text_model(image)
+        return result[0]["generated_text"]
+    
+    except Exception as e:
+        st.error(f"❌ 图片处理失败: {str(e)}")
+        st.stop()
 
 def text2story(text):
     # 优化prompt模板