欢迎访问《图学学报》 分享到：
图学学报

图学学报 ›› 2021, Vol. 42 ›› Issue (6): 883-890.DOI: 10.11996/JG.j.2095-302X.2021060883

• 图像处理与计算机视觉 • 上一篇    下一篇

注意力残差多尺度特征增强的显著性实例分割

  

  1. 华北理工大学人工智能学院，河北 唐山 063210
  • 出版日期:2022-01-18 发布日期:2022-01-18
  • 基金资助:
    国家自然科学基金项目(61502143)；河北省研究生示范课项目(KCJSX2019097)；华北理工大学杰出青年基金项目(JQ201715)；唐山市 人才资助项目(A202110011) 

Salient instance segmentation via attention residual multi-scale feature enhancement  

  1. College of Artificial Intelligence, North China University of Science and Technology, Tangshan Hebei 063210, China
  • Online:2022-01-18 Published:2022-01-18
  • Supported by:
    National Natural Science Foundation of China (61502143); Graduate Model Class Project of Hebei Province (KCJSX2019097); Distinguished Youth Foundation of North China University of Science and Technology (JQ201715); Talent Foundation of Tangshan (A202110011) 

PDF (PC)

197

摘要： 显著性实例分割是指分割出图像中最引人注目的实例对象。现有的显著性实例分割方法中存在 较小显著性实例不易检测分割，以及较大显著性实例分割精度不足等问题。针对这 2 个问题，提出了一种新的 显著性实例分割模型，即注意力残差多尺度特征增强网络(ARMFE)。模型 ARMFE 主要包括 2 个模块：注意力 残差网络模块和多尺度特征增强模块，注意力残差网络模块是在残差网络基础上引入注意力机制，分别从通道 和空间对特征进行选择增强；多尺度特征增强模块则是在特征金字塔基础上进一步增强尺度跨度较大的特征信 息融合。因此，ARMFE 模型通过注意力残差多尺度特征增强，充分利用多个尺度特征的互补信息，同时提升 较大显著性实例对象和较小显著性实例对象的分割效果。ARMFE 模型在显著性实例分割数据集 Salient Instance Saliency-1K (SIS-1K)上进行了实验，分割精度和速度都得到了提升，优于现有的显著性实例分割算法 MSRNet 和 S4Net。

关键词: 显著性实例分割, 注意力机制, 残差网络, 多尺度, 特征增强

Abstract: Salient instance segmentation is to segment the most noticeable instance object in the image. However, there remain some problems in the existing methods of salient instance segmentation. For example, the small salient instances are difficult to be detected and segmented, and the segmentation accuracy is insufficient for large salient instances. Therefore, to solve these two problems, a new salient instance segmentation model, namely the attention residual multi-scale feature enhancement network (ARMFE), has been proposed. ARMFE includes two modules, i.e. the attention residual network module and the multi-scale feature enhancement module. The attention residual network module combines the residual network with the spatial attention sub-module and the channel attention sub-module to enhance the features. The multi-scale feature enhancement module can further enhance the information fusion for features with large scale span based on the feature pyramid. Therefore, the proposed ARMFE model can make full use of the complementary information of multi-scales features by attention residual multi-scale feature enhancement, and then simultaneously improve the accuracy of detecting and segmenting large instance objects and small instance objects. The proposed ARMFE model has been tested on the salient instance segmentation dataset Salient Instance Saliency-1K (SIS-1K), and the segmentation accuracy and speed have been improved. This indicates that our proposed model outperforms other existing salient instance segmentation algorithms, such as MSRNet and S4Net. 

Key words: salient instance segmentation, attention mechanism, residual network, multi-scale, feature enhancement

中图分类号: