Mangaocr Hoogberta V2

🚀 MangaOCR Hoogberta-v2

This is a TrOCR-based image-to-text model that can convert images into text, suitable for scenarios such as manga text recognition.

🚀 Quick Start

Here is how to use this model in PyTorch:

from transformers import TrOCRProcessor, VisionEncoderDecoderModel
from PIL import Image

processor = TrOCRProcessor.from_pretrained('dsupa/mangaocr-hoogberta-v2')
model = VisionEncoderDecoderModel.from_pretrained('dsupa/mangaocr-hoogberta-v2')

def predict(image_path):
    image = Image.open(image_path).convert("RGB")
    pixel_values = processor(images=image, return_tensors="pt").pixel_values
    generated_ids = model.generate(pixel_values)
    generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]

    return generated_text

image_path = "your_img.jpg"
# Bug fix: Change 'predit' to 'predict'
pred = predict(image_path)
print(pred)

💻 Usage Examples

Basic Usage

from transformers import TrOCRProcessor, VisionEncoderDecoderModel
from PIL import Image

processor = TrOCRProcessor.from_pretrained('dsupa/mangaocr-hoogberta-v2')
model = VisionEncoderDecoderModel.from_pretrained('dsupa/mangaocr-hoogberta-v2')

def predict(image_path):
    image = Image.open(image_path).convert("RGB")
    pixel_values = processor(images=image, return_tensors="pt").pixel_values
    generated_ids = model.generate(pixel_values)
    generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]

    return generated_text

image_path = "your_img.jpg"
# Bug fix: Change 'predit' to 'predict'
pred = predict(image_path)
print(pred)

Advanced Usage

# You can expand the model's capabilities by modifying parameters or using loops to process multiple images.
from transformers import TrOCRProcessor, VisionEncoderDecoderModel
from PIL import Image
import os

processor = TrOCRProcessor.from_pretrained('dsupa/mangaocr-hoogberta-v2')
model = VisionEncoderDecoderModel.from_pretrained('dsupa/mangaocr-hoogberta-v2')

def predict(image_path):
    image = Image.open(image_path).convert("RGB")
    pixel_values = processor(images=image, return_tensors="pt").pixel_values
    generated_ids = model.generate(pixel_values)
    generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]

    return generated_text

image_folder = "your_image_folder"
for filename in os.listdir(image_folder):
    if filename.endswith(('.jpg', '.jpeg', '.png')):
        image_path = os.path.join(image_folder, filename)
        pred = predict(image_path)
        print(f"Result for {filename}: {pred}")