Redirigiendo al acceso original de articulo en 18 segundos...
Inicio  /  Applied Sciences  /  Vol: 10 Par: 1 (2020)  /  Artículo
ARTÍCULO
TITULO

Panoptic Segmentation-Based Attention for Image Captioning

Wenjie Cai    
Zheng Xiong    
Xianfang Sun    
Paul L. Rosin    
Longcun Jin and Xinyi Peng    

Resumen

Image captioning is the task of generating textual descriptions of images. In order to obtain a better image representation, attention mechanisms have been widely adopted in image captioning. However, in existing models with detection-based attention, the rectangular attention regions are not fine-grained, as they contain irrelevant regions (e.g., background or overlapped regions) around the object, making the model generate inaccurate captions. To address this issue, we propose panoptic segmentation-based attention that performs attention at a mask-level (i.e., the shape of the main part of an instance). Our approach extracts feature vectors from the corresponding segmentation regions, which is more fine-grained than current attention mechanisms. Moreover, in order to process features of different classes independently, we propose a dual-attention module which is generic and can be applied to other frameworks. Experimental results showed that our model could recognize the overlapped objects and understand the scene better. Our approach achieved competitive performance against state-of-the-art methods. We made our code available.

 Artículos similares

       
 
Qirui Bo, Junwei Liu, Wenchang Shang, Ankit Garg, Xiaoru Jia and Kaiyue Sun    
Nowadays, the use of new compound chemical stabilizers to treat marine clay has gained significant attention. However, the complex non-linear relationship between the influencing factors and the unconfined compressive strength of chemically treated marin... ver más

 
Weilong Guang, Peng Wang, Jinshuai Zhang, Linjuan Yuan, Yue Wang, Guang Feng and Ran Tao    
Predicting the flow situation of cavitation owing to its high-dimensional nonlinearity has posed great challenges. To address these challenges, this study presents a novel reduced order modeling (ROM) method to accurately analyze and predict cavitation f... ver más

 
Yuyang Liu, Bo Feng and Yu Yao    
With the intensification of water pollution problems worldwide, constructed wetlands, as a green, efficient, and energy-saving wastewater treatment technology, have gradually attracted the wide attention of scholars at home and abroad. In order to better... ver más
Revista: Water

 
Jiju Guo, Wengeng Cao, Guohui Lang, Qifa Sun, Tian Nan, Xiangzhi Li, Yu Ren and Zeyan Li    
The presence of high concentrations of geogenic arsenic (As) in groundwater poses a serious threat to the health of millions of individuals globally. This paper examines the research progress of groundwater with high concentrations of geogenic As through... ver más
Revista: Water

 
Mingyu Xie, Xiaoran Zhang, Yuanyuan Jing, Xinyue Du, Ziyang Zhang and Chaohong Tan    
Groundwater is an important part of the water resources, crucial for human production and life. With the rapid development of industry and agriculture, organic pollution of groundwater has attracted great attention. Enhanced in-situ bioremediation of gro... ver más
Revista: Water