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
[21]   Text-Guided Image Manipulation via Generative Adversarial Network With Referring Image Segmentation-Based Guidance [J].
Watanabe, Yuto ;
Togo, Ren ;
Maeda, Keisuke ;
Ogawa, Takahiro ;
Haseyama, Miki .
IEEE ACCESS, 2023, 11 :42534-42545
[22]   Exploring Fine-Grained Image-Text Alignment for Referring Remote Sensing Image Segmentation [J].
Lei, Sen ;
Xiao, Xinyu ;
Zhang, Tianlin ;
Li, Heng-Chao ;
Shi, Zhenwei ;
Zhu, Qing .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2025, 63
[23]   RRSIS: Referring Remote Sensing Image Segmentation [J].
Yuan, Zhenghang ;
Mou, Lichao ;
Hua, Yuansheng ;
Zhu, Xiao Xiang .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2024, 62 :1-12
[24]   A 3D CNN-LSTM-Based Image-to-Image Foreground Segmentation [J].
Akilan, Thangarajah ;
Wu, Qingming Jonathan ;
Safaei, Amin ;
Huo, Jie ;
Yang, Yimin .
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2020, 21 (03) :959-971
[25]   Referring Image Segmentation by Generative Adversarial Learning [J].
Qiu, Shuang ;
Zhao, Yao ;
Jiao, Jianbo ;
Wei, Yunchao ;
Wei, Shikui .
IEEE TRANSACTIONS ON MULTIMEDIA, 2020, 22 (05) :1333-1344
[26]   Image segmentation in marine environments using convolutional LSTM for temporal context [J].
Hansen, Kasper Foss ;
Yao, Linghong ;
Ren, Kang ;
Wang, Sen ;
Liu, Wenwen ;
Liu, Yuanchang .
APPLIED OCEAN RESEARCH, 2023, 139
[27]   Brain Image Segmentation for Ultrascale Neuron Reconstruction via an Adaptive Dual-Task Learning Network [J].
Liu, Min ;
Wu, Shuhan ;
Chen, Runze ;
Lin, Zhuangdian ;
Wang, Yaonan ;
Meijering, Erik .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2024, 43 (07) :2574-2586
[28]   Retinal Vessel Image Segmentation Based on Improved Convolutional Neural Network [J].
Wu Chenyue ;
Yi Benshun ;
Zhang Yungang ;
Huang Song ;
Feng Yu .
ACTA OPTICA SINICA, 2018, 38 (11)
[29]   Multiple Relational Learning Network for Joint Referring Expression Comprehension and Segmentation [J].
Hua, Guoguang ;
Liao, Muxin ;
Tian, Shishun ;
Zhang, Yuhang ;
Zou, Wenbin .
IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 25 :8805-8816
[30]   A survey of methods for addressing the challenges of referring image segmentation [J].
Ji, Lixia ;
Du, Yunlong ;
Dang, Yiping ;
Gao, Wenzhao ;
Zhang, Han .
NEUROCOMPUTING, 2024, 583