[News] We are refactorizing our HAWP as a python package for inference.
- We propose a fast and parsimonious parsing method HAWP to accurately and robustly detect a vectorized wireframe in an input image with a single forward pass.
- The proposed HAWP is fully end-to-end.
- The proposed HAWP does not require the squeeze module.
- State-of-the-art performance on the Wireframe dataset and YorkUrban dataset.
- The proposed HAWP achieves 29.5 FPS on a GPU (Tesla V100) for 1-batch inference.
- Install PyTorch from https://pytorch.org/get-started/locally/
- Install hawp to your python environment by
pip install git+https://github.com/cherubicXN/hawp.git@inference
python -m hawp.predict [options] images
python -m hawp.predict --help
usage: python -m hawp.predict [options] images
positional arguments:
images input images (default: None)
optional arguments:
-h, --help show this help message and exit
--glob GLOB glob expression for input images (for many images)
(default: None)
-o [IMAGE_OUTPUT], --image-output [IMAGE_OUTPUT]
Whether to output an image, with the option to specify
the output path or directory (default: None)
--json-output [JSON_OUTPUT]
Whether to output a json file, with the option to
specify the output path or directory (default: None)
--disable-cuda disable CUDA (default: False)
show:
--show show every plot, i.e., call matplotlib show()
(default: False)
--edge-threshold EDGE_THRESHOLD
show the wireframe edges whose confidences are greater
than [edge_threshold] (default: None)
--out-ext OUT_EXT save the plot in specific format (default: png)
If you find our work useful in your research, please consider citing:
@inproceedings{HAWP,
title = "Holistically-Attracted Wireframe Parsing",
author = "Nan Xue and Tianfu Wu and Song Bai and Fu-Dong Wang and Gui-Song Xia and Liangpei Zhang and Philip H.S. Torr
",
booktitle = "IEEE Conference on Computer Vision and Pattern Recognition (CVPR)",
year = {2020},
}
-
We acknowledge the effort from the authors of the Wireframe dataset and the YorkUrban dataset. These datasets make accurate line segment detection and wireframe parsing possible.
-
This branch is partially based on openpifpaf.