arrow
返回

Multi-Grained Representation Aggregating Transformer with Gating Cycle for Change Captioning

delete2024-09-12
delete1
delete
OA
AI
S
Shengbin Yue
Y
Yunbin Tu
李
李亮 (Liang Li)
高盛祥 封面图
高盛祥 (Shengxiang Gao) *
余正涛 封面图
余正涛 (Zhengtao Yu)
DOI:10.1145/3660346delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Change captioning aims to describe the difference within an image pair in natural language, which combines visual comprehension and language generation. Although significant progress has been achieved, it remains a key challenge of perceiving the object change from different perspectives, especially the severe situation with drastic viewpoint change. In this article, we propose a novel full-attentive network, namely Multi- grained Representation Aggregating Transformer (MURAT), to distinguish the actual change from viewpoint change. Specifically, the Pair Encoder first captures similar semantics between pairwise objects in a multilevel manner, which are regarded as the semantic cues of distinguishing the irrelevant change. Next, a novel Multi-grained Representation Aggregator (MRA) is designed to construct the reliable difference representation by employing both coarse- and fine-grained semantic cues. Finally, the language decoder generates a description of the change based on the output of MRA. Besides, the Gating Cycle Mechanism is introduced to facilitate the semantic consistency between difference representation learning and language generation with a reverse manipulation, so as to bridge the semantic gap between change features and text features. Extensive experiments demonstrate that the proposed MURAT can greatly improve the ability to describe the actual change in the distraction of irrelevant change and achieves state-of-the-art performance on three benchmarks, CLEVR-Change, CLEVR-DC, and Spot-the-Diff.
Keyword:
Change captioning
multi-grained representation aggregating
gating cycle
Transformer

期刊

ACM Transactions on Multimedia Computing Communications and Applications 封面图
ACM Transactions on Multimedia Computing Communications and Applications
IF:
6
论文数:
2.0K
被引数:
5.4K

机构

U
university of chinese academy of sciences, cas
学者数:
4.1W
论文数: 3.8W
被引数: 75
I
institute of computing technology, cas
学者数:
1.0K
论文数: 878
被引数: 1
C
chinese academy of sciences
学者数:
56.7W
论文数: 45.0W
被引数: 704
学者 查看更多机构
引用论文

引用论文

Expression in baculovirus vector system of the nucleocapsid protein gene of rinderpest virus
err1993-07-01
err0
PREAI
errH. Kamata; S. Ohkubo; M. Sugiyama; Y. Matsuura; Y. Kamata; K. Tsukiyama-Kohara; K. Imaoka; C. Kai; Y. Yoshikawa; K. Yamanouchi
err分享
err收藏
Viewpoint-Adaptive Representation Disentanglement Network for Change Captioning
err2023-01-01
err11
PREAI
errTu, Yunbin; Li, Liang; Su, Li; Du, Junping; Lu, Ke; Huang, Qingming
err分享
err收藏
Remote sensing of fish-processing in the Sundarbans Reserve Forest, Bangladesh: an insight into the modern slavery-environment nexus in the coastal fringe
err2020-09-17
err0
errOAAI
errBethany Jackson; Doreen S. Boyd; Christopher D. Ives; Jessica L. Decker Sparks; Giles M. Foody; Stuart Marsh; Kevin Bales
err分享
err收藏
A Hierarchical CNN-RNN Approach for Visual Emotion Classification
err2019-12-07
err22
PREAI
errLi, Liang; Zhu, Xinge; Hao, Yiming; Wang, Shuhui; Gao, Xingyu; Huang, Qingming
err分享
err收藏
Portal Vein and Hepatic Vein Segmentation in Multi-Phase MR Images Using Flow-Guided Change Detection
err2022-01-01
err10
PREAI
errGuo, Qing; Song, Hong; Fan, Jingfan; Ai, Danni; Gao, Yuanjin; Yu, Xiaoling; Yang, Jian
err分享
err收藏
Hierarchical Paired Channel Fusion Network for Street Scene Change Detection
err2021-01-01
err58
errOAAI
errLei, Yinjie; Peng, Duo; Zhang, Pingping; Ke, Qiuhong; Li, Haifeng
err分享
err收藏
学者 查看更多内容