Multimodal genre prediction using OCR

Check out our report here!

CRAFT text extraction code:

https://github.com/clovaai/CRAFT-pytorch

https://github.com/clovaai/deep-text-recognition-benchmark

Install dependencies

Requirements

• PyTorch>=0.4.1
• torchvision>=0.2.1
• opencv-python>=3.4.2
• Flask >= 1.1.2
• requirements.txt

pip install -r requirements.txt

Instructions

Load pre-trained weights

• Insert craft_mlt_25k.pth into the CRAFT-pytorch folder
• Insert TPS-ResNet-BiLSTM-Attn.pth into deep-text-recognition-benchmark folder
• Insert best_gmu.pth, best_model_mmu, and best_img_model into genre-demo folder

Run demo:

cd genre-demo
python3 main.py