READMEori.md

Difference (Written by Una 07/31/2020)

• Add train codes (uses CrossEntropy according to the paper)
• "AverageMeter" refer to InnovArul/pytorch_utils

---

Code for paper "Deep Hough Transform for Semantic Line Detection" (ECCV2020).

Deep Hough Transform

Requirements

numpy
scipy
opencv-python
scikit-image
pytorch>=1.0
torchvision
tqdm
yml
deep-hough

To install deep-hough, run the following commands.

cd deep-hough-transform
cd model/_cdht
python setup.py build 
python setup.py install --user

Pretrain model (based on ResNet50-FPN): https://drive.google.com/file/d/1a6Rbu1Bslyo9sjNlUUdi7NnSTdRIWwS5/view?usp=sharing

Forward

Generate visualization results and save coordinates to _.npy file.

CUDA_VISIBLE_DEVICES=0 python forward.py --model （your_best_model.pth） --tmp (your_result_save_dir)

Test

Test the EA-score on SEL dataset. After forwarding the model and get the coordinates files. Run the following command to produce EA-score.

python test.py --pred result/debug/visualize_test/(change to your onw path which includes _.npy files) --gt gt_path/include_txt