Update app.py

app.py

@@ -2,7 +2,7 @@ import os
 from huggingface_hub import login
 from transformers import BlipProcessor, BlipForConditionalGeneration
 from PIL import Image
-import pytesseract
+import easyocr
 
 
 
@@ -31,11 +31,13 @@ model.to(device)
 @spaces.GPU(duration=300)
 def generate_caption_and_image(image):
     img = image.convert("RGB")
-
+    reader = easyocr.Reader(['en'])  
+    result = reader.readtext(img)
     
-    extracted_text = pytesseract.image_to_string(img)
-    # Process the image
-    raw_image = image.convert("RGB")
+    # Print the extracted text
+    for detection in result:
+        print(f"Detected text: {detection[1]}")
+
       
 
     
@@ -44,7 +46,7 @@ def generate_caption_and_image(image):
     inputs = {key: val.to(device) for key, val in inputs.items()}
     out = model.generate(**inputs)
     caption = processor.decode(out[0], skip_special_tokens=True)
-    prompt = f"Create a highly realistic design of a clothing item based on the following description: 'The design should incorporate elements from the extracted text: {extracted_text}, along with the caption: {caption}. The clothing should look realistic, modern, and stylish. Use high-quality fabric textures and realistic lighting to give the design a lifelike appearance. The colors, patterns, and materials should reflect the essence of the caption and extracted text.'"
+    prompt = f"Create a highly realistic design of a clothing item based on the following description: 'The design should incorporate elements from the extracted text: {result}, along with the caption: {caption}. The clothing should look realistic, modern, and stylish. Use high-quality fabric textures and realistic lighting to give the design a lifelike appearance. The colors, patterns, and materials should reflect the essence of the caption and extracted text.'"
 
 
     # Generate image based on the caption