Dual Convolutional LSTM Network for Referring Image Segmentation

被引:39
作者
Ye, Linwei [1 ]
Liu, Zhi [2 ,3 ]
Wang, Yang [1 ]
机构
[1] Univ Manitoba, Dept Comp Sci, Winnipeg, MB R3T 2N2, Canada
[2] Shanghai Univ, Shanghai Inst Adv Commun & Data Sci, Shanghai 200444, Peoples R China
[3] Shanghai Univ, Sch Commun & Informat Engn, Shanghai 200444, Peoples R China
基金
中国国家自然科学基金; 加拿大自然科学与工程研究理事会;
关键词
Image segmentation; Visualization; Decoding; Linguistics; Task analysis; Logic gates; Computer vision; Referring image segmentation; encoder-decoder; vision and language; deep learning;
D O I
10.1109/TMM.2020.2971171
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We consider referring image segmentation. It is a problem at the intersection of computer vision and natural language understanding. Given an input image and a referring expression in the form of a natural language sentence, the goal is to segment the object of interest in the image referred by the linguistic query. To this end, we propose a dual convolutional LSTM (ConvLSTM) network to tackle this problem. Our model consists of an encoder network and a decoder network, where ConvLSTM is used in both encoder and decoder networks to capture spatial and sequential information. The encoder network extracts visual and linguistic features for each word in the expression sentence, and adopts an attention mechanism to focus on words that are more informative in the multimodal interaction. The decoder network integrates the features generated by the encoder network at multiple levels as its input and produces the final precise segmentation mask. Experimental results on four challenging datasets demonstrate that the proposed network achieves superior segmentation performance compared with other state-of-the-art methods.
引用
收藏
页码:3224 / 3235
页数:12
相关论文
共 50 条
[41]   Residual Convolutional Neural Network for Cardiac Image Segmentation and Heart Disease Diagnosis [J].
Liu, Tao ;
Tian, Yun ;
Zhao, Shifeng ;
Huang, Xiaoying ;
Wang, Qingjun .
IEEE ACCESS, 2020, 8 :82153-82161
[42]   UVI Image Segmentation of Auroral Oval: Dual Level Set and Convolutional Neural Network Based Approach [J].
Tian, Chenjing ;
Du, Huadong ;
Yang, Pinglv ;
Zhou, Zeming ;
Weng, Libin .
APPLIED SCIENCES-BASEL, 2020, 10 (07)
[43]   Dual-Branch Network for Cloud and Cloud Shadow Segmentation [J].
Lu, Chen ;
Xia, Min ;
Qian, Ming ;
Chen, Binyu .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
[44]   A Hybrid Framework for Referring Image Segmentation: Dual-Decoder Model with SAM Complementation [J].
Chen, Haoyuan ;
Zhou, Sihang ;
Li, Kuan ;
Yin, Jianping ;
Huang, Jian .
MATHEMATICS, 2024, 12 (19)
[45]   Instance-Specific Feature Propagation for Referring Segmentation [J].
Liu, Chang ;
Jiang, Xudong ;
Ding, Henghui .
IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 25 :3657-3667
[46]   Language-Aware Vision Transformer for Referring Segmentation [J].
Yang, Zhao ;
Wang, Jiaqi ;
Ye, Xubing ;
Tang, Yansong ;
Chen, Kai ;
Zhao, Hengshuang ;
Torr, Philip H. S. .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2025, 47 (07) :5238-5255
[47]   Cross-Modal Progressive Comprehension for Referring Segmentation [J].
Liu, Si ;
Hui, Tianrui ;
Huang, Shaofei ;
Wei, Yunchao ;
Li, Bo ;
Li, Guanbin .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2022, 44 (09) :4761-4775
[48]   CSUnet: a dual attention and hybrid convolutional network for polyp segmentation [J].
Liu, Shangwang ;
Si, Feiyan ;
Lin, Yinghai .
SIGNAL IMAGE AND VIDEO PROCESSING, 2024, 18 (11) :8445-8456
[49]   Vision-Aware Language Reasoning for Referring Image Segmentation [J].
Xu, Fayou ;
Luo, Bing ;
Zhang, Chao ;
Xu, Li ;
Pu, Mingxing ;
Li, Bo .
NEURAL PROCESSING LETTERS, 2023, 55 (08) :11313-11331
[50]   Vision-Aware Language Reasoning for Referring Image Segmentation [J].
Fayou Xu ;
Bing Luo ;
Chao Zhang ;
Li Xu ;
Mingxing Pu ;
Bo Li .
Neural Processing Letters, 2023, 55 :11313-11331