Update app.py

app.py

@@ -2,35 +2,49 @@ import gradio as gr
 import torch
 import pytesseract
 from transformers import AutoTokenizer, AutoModelForCausalLM
+import cv2  # Ensure you have OpenCV installed
 
 # Load models and tokenizers
-tokenizer_eng = AutoTokenizer.from_pretrained("ucaslcl/GOT-OCR2_0")
-tokenizer_hin = AutoTokenizer.from_pretrained("ucaslcl/GOT-OCR2_0")
-model_eng = AutoModelForCausalLM.from_pretrained("ucaslcl/GOT-OCR2_0")
-model_hin = AutoModelForCausalLM.from_pretrained("ucaslcl/GOT-OCR2_0")
+tokenizer_eng = AutoTokenizer.from_pretrained("ucaslcl/GOT-OCR2_0", trust_remote_code=True)
+tokenizer_hin = AutoTokenizer.from_pretrained("ucaslcl/GOT-OCR2_0", trust_remote_code=True)
+model_eng = AutoModelForCausalLM.from_pretrained("ucaslcl/GOT-OCR2_0", trust_remote_code=True)
+model_hin = AutoModelForCausalLM.from_pretrained("ucaslcl/GOT-OCR2_0", trust_remote_code=True)
 
-def perform_ocr(image, language):
+def perform_ocr(image):
+    """Perform OCR on the image for both English and Hindi."""
     # Set device to CPU or GPU
     device = "cuda" if torch.cuda.is_available() else "cpu"
     
     model_eng.to(device)
     model_hin.to(device)
 
-    # Process the image using the appropriate model
-    img_path = "path/to/your/image"  # Set this path appropriately
-    res_eng = model_eng.chat(tokenizer_eng, img_path, ocr_type='ocr')
-    img_cv = cv2.imread(img_path)
-    
+    # Convert the input image to an OpenCV format
+    img_cv = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)  # Convert to BGR for OpenCV
+    img_path = "temp_image.jpg"  # Temporary path to save the image
+    cv2.imwrite(img_path, img_cv)  # Save the image temporarily
+
+    # Use pytesseract for English OCR
+    english_text = pytesseract.image_to_string(img_cv)
+
     # Use pytesseract for Hindi OCR
     tesseract_config = '--psm 6'
-    res_hin = pytesseract.image_to_string(img_cv, lang='hin', config=tesseract_config)
+    hindi_text = pytesseract.image_to_string(img_cv, lang='hin', config=tesseract_config)
 
-    return res_eng, res_hin
+    return english_text, hindi_text
 
-def ocr_and_search(image, language):
-    english_text, hindi_text = perform_ocr(image, language)
+def ocr_and_search(image):
+    """Process the image and extract text in both languages."""
+    english_text, hindi_text = perform_ocr(image)
     return english_text, hindi_text
 
-# Gradio
-iface = gr.Interface(fn=ocr_and_search, inputs=["image", "dropdown"], outputs=["text", "text"])
+# Gradio interface
+iface = gr.Interface(
+    fn=ocr_and_search,
+    inputs=gr.inputs.Image(type="numpy"),  # Use numpy array for OpenCV
+    outputs=["text", "text"],
+    title="OCR for English and Hindi",
+    description="Upload an image to extract text in English and Hindi."
+)
+
+# Launch the interface
 iface.launch()