Pytorch-yolo-phoc

Implementation on pytorch of the code from the ECCV 2018 paper - Single Shot Scene Text Retrieval. Paper: https://arxiv.org/abs/1808.09044

This code uses the YOLOv2 implementation from https://github.com/marvis/pytorch-yolo2 and modifies it respectively.

Training YOLO-PHOC

---

All paths are hardcoded and need to be edited accordingly.

Modify Cfg files for training

Change the cfg/XXXX.data file according to training objective

train  = path_to_file_with_list_of_files_to_train.txt
names = data/recognition.names
backup = backup
gpus  = 0
num_workers = 10

The file cfg/XXXX.cfg contains the config parameters for training.

A folder/file needs to be specified with the images for training time.

Download Pretrained Convolutional Weights

Download weights from the convolutional layers (Imagenet pre-trained weights)

wget http://pjreddie.com/media/files/darknet19_448.conv.23

Train The Model

Modify the options in train.py file.

python train.py

---

Detection Using A Pre-Trained Model

---

The model has been trained, achieving the following results:

• IIIT Scene Text Retrieval dataset: 67.26
• IIIT Sports-10k dataset: 72.73
• Street View Text (SVT) dataset: 83.14

The weights can be downloaded from: https://drive.google.com/file/d/1nmlzLJNPvI4Mw2bB6t-1N29tBz9kLb4t/view?usp=sharing