Joint Visual Context for Pedestrian Captioning | |
Quan Liu1,2,3; Sijiong Zhang1,2,3 | |
2018 | |
Conference Name | 9th International Conference, ICIMCS 2017 |
Source Publication | Internet Multimedia Computing and Service((CCIS, volume 819) |
Conference Date | 2017-8-23 |
Conference Place | Qingdao, China |
Publication Place | Switzerland |
Publisher | Springer |
Abstract |
Image captioning is a fundamental task connecting computer vision and natural language processing. Recent researches usually concentrate on generic image captioning or video captioning among thousands of classes. However, they can not effectively deal with a specific class of objects, such as pedestrian. Pedestrian captioning is critical for analysis, identification and retrieval in massive collections of data. Therefore, in this paper, we propose a novel approach for pedestrian captioning with joint visual context. Firstly, a deep convolutional neural network (CNN) is employed to obtain the global attributes of a pedestrian (e.g., gender, age, and actions), and a Faster R-CNN is utilized to detect the local parts of interest for identification of the local attributes of a pedestrian (e.g., cloth type, color type, and the belongings). Then, we splice the global and local attributes into a fixed length vector and input it into a Long-Short Term Memory network (LSTM) to generate descriptions. Finally, a dataset of 5000 pedestrian images is collected to evaluate the performance of pedestrian captioning. Experimental results show the superiority of the proposed approach. |
Keyword | Image Captioning Pedestrian Description |
Subject Area | 天文技术与方法 |
Language | 英语 |
Document Type | 会议论文 |
Identifier | http://ir.niaot.ac.cn/handle/114a32/1534 |
Collection | 会议论文 |
Affiliation | 1.南京天文光学技术研究所 2.天文光学技术重点实验室 3.中国科学院大学 |
Recommended Citation GB/T 7714 | Quan Liu,Sijiong Zhang. Joint Visual Context for Pedestrian Captioning[C]. Switzerland:Springer,2018. |
Files in This Item: | Download All | |||||
File Name/Size | DocType | Version | Access | License | ||
Joint Visual Context(815KB) | 会议论文 | 开放获取 | CC BY-NC-SA | View Download |
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment