Update app.py

app.py

@@ -2,6 +2,7 @@ import streamlit as st
 import torch
 from PIL import Image
 import gc
+import tempfile
 import os
 from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
 from qwen_vl_utils import process_vision_info
@@ -42,9 +43,14 @@ if image:
     st.write("Extracting text from image...")
     byaldi_model = load_byaldi_model()
     
+    # Save the image to a temporary file
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".jpg") as temp_file:
+        img.save(temp_file, format="JPEG")
+        temp_file_path = temp_file.name
+
     # Create a temporary index for the uploaded image
     with st.spinner("Processing image..."):
-        byaldi_model.index(img, index_name="temp_index", overwrite=True)
+        byaldi_model.index(temp_file_path, index_name="temp_index", overwrite=True)
     
     # Perform a dummy search to get the OCR results
     ocr_results = byaldi_model.search("Extract all text from the image", k=1)
@@ -62,6 +68,9 @@ if image:
     del byaldi_model
     clear_memory()
 
+    # Remove the temporary file
+    os.unlink(temp_file_path)
+
     # Text input field for question
     question = st.text_input("Ask a question about the image and extracted text")