INFNet :Deep instance feature chain learning network for panoptic segmentation

被引：0
|
作者：
Mao L. [1 ]
Ren F.-Z. [1 ]
Yang D.-W. [1 ]
Zhang R.-B. [1 ]
机构：
[1] School of Electromechanical Engineering, Dalian Minzu University, Dalian
来源：
Guangxue Jingmi Gongcheng/Optics and Precision Engineering | 2020年 / 28卷 / 12期
关键词：
Chain network; Edge feature; Instance feature; Panoptic segmentation; Shortcut connection;
D O I：
10.37188/OPE.20202812.2665
中图分类号：
学科分类号：
摘要：
A novel deep instance feature chain learning network for panoptic segmentation (INFNet) was developed to solve the problem of failure of target boundary segmentation caused by insufficient instant feature extraction in panoptic segmentation. This network consisted of a basic chain unit, whose functions were divided into two types, feature holding chain and feature enhancement chain, based on the different methods of processing feature information by the unit structure. The feature-holding chain represented the input stage of the extraction of a chain network feature, in which the integrity of the input information was guaranteed, and then this feature was transmitted to the feature-enhancement chain structure. The feature-enhancement chain increased the network depth and improved the feature extraction ability through its extension. INFNet could obtain adequate edge feature information and improve segmentation accuracy, owing to the robust depth-stacking characteristics. The experiment results for the MS COCO and Cityscapes datasets showed that our INFNet was superior to similar existing methods in terms of segmentation accuracy. Compared to the Mask RCNN instance segmentation structure widely used in panoptic segmentation networks, the segmentation accuracy of INFNet increased by up to 0.94%. © 2020, Science Press. All right reserved.
引用
收藏
页码：2665 / 2673
页数：8
相关论文
共 21 条
  • [11] LI Y, CHEN X, ZHU Z, Et al., Attention-Guided Unified Network for Panoptic Segmentation, IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7019-7028, (2019)
  • [12] ZEILER M D, FERGUS R., Visualizing and Understanding Convolutional Neural Networks, European Conference on Computer Vision, pp. 818-833, (2014)
  • [13] TAN M, LE Q V., EfficientNet: Rethinking model scaling for convolutional neural networks, International Conference on Machine Learning, pp. 6105-6114, (2019)
  • [14] RIPLEY B D., Pattern Recognition and Neural Networks, (1996)
  • [15] NAIR V, HINTON G E., Rectified Linear Units Improve Restricted Boltzmann Machines, International Conference on Machine Learning, pp. 807-814, (2010)
  • [16] LIN T Y, MAIRE M, ELONGIE SB, Et al., Microsoft COCO: Common Objects in Context, European Conference on Computer Vision, pp. 740-755, (2014)
  • [17] CORDTS M, OMRAN M, RANOS S, Et al., The Cityscapes Dataset for Semantic Urban Scene Understanding, IEEE Conference on Computer Vision and Pattern Recognition, pp. 3213-3223, (2016)
  • [18] WANG J L, FU X S, HUANG ZH CH, Et al., Multi-type cooperative targets detection using improved YOLOv2 convolutional neural network, Opt. Precision Eng, 28, 1, pp. 251-260, (2020)
  • [19] DONG X X, HE X H, WU X H, Et al., Object detection algorithm based on attention mask fusion, Chinese Journal of Liquid Crystals and Displays, 34, 8, pp. 825-833, (2019)
  • [20] KIRILLOV A, HE K, GIRSHICK R, Et al., Panoptic Segmentation, IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 9396-9405, (2019)
123