Official implementation of the paper:
STEdge: Self-training Edge Detection with Multi-layer Teaching and Regularization[arXiv].
Yunfan Ye, Renjiao Yi, Zhiping Cai, Kai Xu.
- [Aug 2023] Initial release of code.
Learning-based edge detection has hereunto been strongly supervised with pixel-wise annotations which are tedious to obtain manually. We study the problem of self-training edge detection, leveraging the untapped wealth of large-scale unlabeled image datasets. We design a self-supervised framework with multi-layer regularization and self-teaching. In particular, we impose a consistency regularization which enforces the outputs from each of the multiple layers to be consistent for the input image and its perturbed counterpart. We adopt L0-smoothing as the 'perturbation' to encourage edge prediction lying on salient boundaries following the cluster assumption in self-supervised learning. Meanwhile, the network is trained with multi-layer supervision by pseudo labels which are initialized with Canny edges and then iteratively refined by the network as the training proceeds. The regularization and self-teaching together attain a good balance of precision and recall, leading to a significant performance boost over supervised methods, with lightweight refinement on the target dataset. Furthermore, our method demonstrates strong cross-dataset generality. For example, it attains 4.8% improvement for ODS and 5.8% for OIS when tested on the unseen BIPED dataset, compared to the state-of-the-art methods.
This code has been tested with Ubuntu 18.04, one 3080Ti GPU with CUDA 11.4, Python 3.8, Pytorch 1.12.
Ealier versions may also work.
If you are in a hurry to use the code, please feel free to contact me if you meet any problem. I will organize the code after finishing my current and urgent project. Thanks for your patience.
Download the DexiNed model from Google Drive, which is pre-trained only on COCO-val dataset with Canny [300, 400], and and then run the following code for the self-training process:
python self_train.py@ARTICLE{10187181,
author={Ye, Yunfan and Yi, Renjiao and Cai, Zhiping and Xu, Kai},
journal={IEEE Transactions on Neural Networks and Learning Systems},
title={STEdge: Self-Training Edge Detection With Multilayer Teaching and Regularization},
year={2023},
volume={},
number={},
pages={1-11},
doi={10.1109/TNNLS.2023.3292905}}