Deeply Supervised Salient Object Detection with Short Connections

Qibin Hou1 Ming-Ming Cheng1  Xiaowei Hu1  Ali Borji Zhuowen TuPhilip H. S. Torr4

1CCCE, Nankai University      2CRCV, UCF     3UCSD     4The University of Oxford

Abstract

Recent progress on saliency detection is substantial, benefiting mostly from the explosive development of Convolutional Neural Networks (CNNs). Semantic segmentation and saliency detection algorithms developed lately have been mostly based on Fully Convolutional Neural Networks (FCNs). There is still a large room for improvement over the generic FCN models that do not explicitly deal with the scale-space problem. Holistically-Nested Edge Detector (HED) provides a skip-layer structure with deep supervision for edge and boundary detection, but the performance gain of HED on saliency detection is not obvious. In this paper, we propose a new saliency method by introducing short connections to the skip-layer structures within the HED architecture. Our framework provides rich multi-scale feature maps at each layer, a property that is critically needed to perform segment detection. Our method produces state-of-the-art results on 5 widely tested salient object detection benchmarks, with advantages in terms of efficiency (0.08 seconds per image), effectiveness, and simplicity over the existing algorithms.

Paper

  • Deeply supervised salient object detection with short connections, Qibin Hou, Ming-Ming Cheng, Xiaowei Hu, Zhuowen Tu, Ali Borji, IEEE CVPR, 2017. [pdf] [Project Page] [bib] [source code]

Source Code

You can find our code here. We have uploaded the caffe and CRF packages we used in our paper.

If you find our work is helpful, please cite

@article{hou2016deeply,
  title={Deeply supervised salient object detection with short connections},
  author={Hou, Qibin and Cheng, Ming-Ming and Hu, Xiaowei and Borji, Ali and Tu, Zhuowen and Torr, Philip},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  year={2017}
}

Contact

andrewhoux AT gmail DOT com

Applications

This algorithm is used in flaghship products such as Huawei Mate 10, Hawei Honour V10 etc, to create smart bokeh effects for perfect portraits. See the corresponding

说点什么

18 Comments on "Deeply Supervised Salient Object Detection with Short Connections"

Notify of
avatar
inkfish
Guest
inkfish

您好,我想重新训练您的网络,然而找不到MSRA-B数据集,微软上的下载链接已经失效了,您可不可以提供一个MSRA-B的下载链接

Ming-Ming Cheng
Admin

在我们2015年IEEE TIP 的 Benchmark论文主页能找到所有相关述数据集的下载(百度网盘)。

Wan Yuqi
Guest
Wan Yuqi

Hi, Qibin. When i train the model, i can’t solve the problem ” Unknown layer type: ImageLabelmapData “. If you know a method to solve the problem, please to help me. Thank you very much!!!

Chen
Guest
Chen

I met the same problem.Do you have the method to solve it right now?

wsw
Guest
wsw

您好,我看了您的这篇论文。有个地方没明白。在3.3Inference一元项的定义中,分母中包含sigmoid函数。请问x的取值范围是{0,1}吗,那个h(x)的值域就是{0.5,e/(e+1)},可以这样理解吗

flyer
Guest
flyer

您的hed编译成功了吗?

flyer
Guest
flyer

请教一下您是如何编译的hed提供的caffe的?

flyer
Guest
flyer

您好,我编译hed提供的caffe时,注释了USE_CUDNN=1(因在其他网页看到编译这个需要cuda4,但我的是cudnn5),然后make all 时出错,提示cublas.h_v2.h:No such file or directory.