Spaces:
Sleeping
Sleeping
File size: 3,032 Bytes
763589d 30a00ea c42cef3 763589d e3dbfa9 f90e854 6d15eb6 763589d e3dbfa9 eed28b8 e3dbfa9 f90e854 763589d e3dbfa9 6837700 763589d e3dbfa9 763589d e3dbfa9 763589d e3dbfa9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 |
from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
from qwen_vl_utils import process_vision_info
import streamlit as st
import torch
from PIL import Image
# Default: Load the model on the available device(s)
@st.cache_resource
def init_qwen_model():
_model = Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", torch_dtype="auto", device_map="auto")
_processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")
return _model, _processor
# Modified function to use only the image as the argument
@st.cache_data
def get_qwen_text(uploaded_file, model, processor):
if uploaded_file is not None:
# Open the uploaded image file
image = Image.open(uploaded_file)
st.image(image, caption="Uploaded Image", use_column_width=True)
messages = [
{
"role": "user",
"content": [
{
"type": "image",
"image": image,
},
{"type": "text", "text": "Run Optical Character recognition on the image."},
],
}
]
# Preparation for inference
text = processor.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
image_inputs, video_inputs = process_vision_info(messages)
inputs = processor(
text=[text],
images=image_inputs,
videos=video_inputs,
padding=True,
return_tensors="pt",
)
inputs = inputs.to("cpu")
# Inference: Generation of the output
generated_ids = model.generate(**inputs, max_new_tokens=128)
generated_ids_trimmed = [
out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]
output_text = processor.batch_decode(
generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
return output_text
# Streamlit app title
st.title("OCR Image Text Extraction")
# Initialize the model and processor
MODEL, PROCESSOR = init_qwen_model()
# File uploader for images
uploaded_file = st.file_uploader("Choose an image...", type=["png", "jpg", "jpeg"])
if uploaded_file:
st.subheader("Extracted Text:")
output = get_qwen_text(uploaded_file, MODEL, PROCESSOR)
st.write(output)
# Keyword search functionality
st.subheader("Keyword Search")
search_query = st.text_input("Enter keywords to search within the extracted text")
if search_query:
# Check if the search query is in the extracted text
if search_query.lower() in output.lower():
highlighted_text = output.replace(search_query, f"**{search_query}**")
st.write(f"Matching Text: {highlighted_text}")
else:
st.write("No matching text found.")
else:
st.info("Please upload an image to extract text.") |