DBRS2: dense boundary regression for semantic segmentation

DBRS2: dense boundary regression for semantic segmentation
复制标题

DBRS2:用于语义分割的密集边界回归

DOI:
10.1117/1.jei.27.5.053033
复制
发表时间:
2018-10
影响因子:
1.1
通讯作者:
Meijie Wang
Meijie Wang
中科院分区:
计算机科学4区
文献类型:
--
作者:
Jinfu Yang;Jingling Zhang;Mingai Li;Meijie Wang

文献摘要

参考文献

相似文献

抽象。目前大多数语义分割方法都依赖于完全卷积网络实现了最先进的性能。然而,诸如池化或卷积跨越的连续操作导致空间上不连贯的对象边界。我们提出了一个密集的边界回归架构(DBRS 2),其目的是使用边界线索,以帮助高层次的语义分割任务。具体来说,我们首先提出了一种多级引导的低级边界(MG-LB)学习方法,其中我们利用多级卷积特征作为低级边界检测的指导。预测的MG-LB边界用于实现一致的空间分组并增强对段边界的精确遵守。然后,我们提出了一个基于边界惩罚和外观惩罚的全局能量模型,它们分别定义在DeepLabv 3网络获得的预测边界和粗分割上。最后,通过最小化全局能量模型来回归细化分割。在PASCAL VOC 2012、ADE 20 K、CamVid和BSD 500数据集上进行的大量实验表明,该方法在语义分割和边界检测任务上都可以获得最先进的性能。
Abstract. Most of the current semantic segmentation approaches have achieved state-of-the-art performance relying on fully convolutional networks. However, the consecutive operations such as pooling or convolution striding lead to spatially disjointed object boundaries. We present a dense boundary regression architecture (DBRS2), which aims to use boundary cues to aid high-level semantic segmentation task. Specifically, we first propose a multilevel guided low-level boundary (MG-LB) learning method, where we exploit multilevel convolutional features as guidance for low-level boundary detection. The predicted MG-LB boundaries are used to enable consistent spatial grouping and enhance precise adherence to segment boundaries. Then, we present a significant global energy model based on boundary penalty and appearance penalty, which are respectively defined on the predicted boundaries and coarse segmentations obtained by the DeepLabv3 network. Finally, the refined segmentations are regressed by minimizing the global energy model. Extensive experiments over PASCAL VOC 2012, ADE20K, CamVid, and BSD500 datasets demonstrate that the proposed approach can obtain state-of-the-art performance on both semantic segmentation and boundary detection tasks.
DOI: --
发表时间: 2012-12
期刊: --
影响因子: --
作者:
Xiaofeng Ren;Liefeng Bo
通讯作者: Xiaofeng Ren;Liefeng Bo
DOI: 10.1007/s11263-018-1140-0
发表时间: 2019-03-01
影响因子: 19.5
作者:
Zhou, Bolei;Zhao, Hang;Torralba, Antonio
通讯作者: Torralba, Antonio
DOI: 10.1109/tpami.2016.2572683
发表时间: 2017-04-01
影响因子: 23.6
作者:
Shelhamer, Evan;Long, Jonathan;Darrell, Trevor
通讯作者: Darrell, Trevor
DOI: 10.1109/cvpr.2004.173
发表时间: 2004-06
期刊: Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004.
影响因子: --
作者:
Xuming He;R. Zemel;M. A. Carreira-Perpiñán
通讯作者: Xuming He;R. Zemel;M. A. Carreira-Perpiñán
DOI: 10.1109/iccv.2011.6126219
发表时间: 2011-11
期刊: 2011 International Conference on Computer Vision
影响因子: --
作者:
Aurélien Lucchi;Yunpeng Li;X. Boix;Kevin Smith;P. Fua
通讯作者: Aurélien Lucchi;Yunpeng Li;X. Boix;Kevin Smith;P. Fua