English

Learning to Refine Object Contours with a Top-Down Fully Convolutional Encoder-Decoder Network

Computer Vision and Pattern Recognition 2017-07-13 v1

Abstract

We develop a novel deep contour detection algorithm with a top-down fully convolutional encoder-decoder network. Our proposed method, named TD-CEDN, solves two important issues in this low-level vision problem: (1) learning multi-scale and multi-level features; and (2) applying an effective top-down refined approach in the networks. TD-CEDN performs the pixel-wise prediction by means of leveraging features at all layers of the net. Unlike skip connections and previous encoder-decoder methods, we first learn a coarse feature map after the encoder stage in a feedforward pass, and then refine this feature map in a top-down strategy during the decoder stage utilizing features at successively lower layers. Therefore, the deconvolutional process is conducted stepwise, which is guided by Deeply-Supervision Net providing the integrated direct supervision. The above proposed technologies lead to a more precise and clearer prediction. Our proposed algorithm achieved the state-of-the-art on the BSDS500 dataset (ODS F-score of 0.788), the PASCAL VOC2012 dataset (ODS F-score of 0.588), and and the NYU Depth dataset (ODS F-score of 0.735).

Keywords

Cite

@article{arxiv.1705.04456,
  title  = {Learning to Refine Object Contours with a Top-Down Fully Convolutional Encoder-Decoder Network},
  author = {Yahui Liu and Jian Yao and Li Li and Xiaohu Lu and Jing Han},
  journal= {arXiv preprint arXiv:1705.04456},
  year   = {2017}
}

Comments

12 pages, 13 figures