科研项目

项目列表


Video Content Annotation


视频内容标注


Introduction

   

Video Annotation via Image Groups from the Web

 

Searching desirable events in uncontrolled videos is a challenging task. It is time consuming and labor expensive to collect a large amount of required labeled videos for training event models under various circumstances. To alleviate this problem, we propose to leverage abundant Web images for videos since Web images contain a rich source of information with many events roughly annotated and taken under various conditions. However, knowledge from theWeb is noisy and diverse, brute force knowledge transfer of imagesmay hurt the video annotation performance. Therefore, we propose a novel Groupbased Domain Adaptation (GDA) learning framework to leverage different groups of knowledge (source domain) queried from the Web image search engine to consumer videos (target domain).

 

Figure

 

 

Papers

1. Han Wang, Xinxiao Wu and Yunde Jia. Video Annotation via Image Groups from the Web. IEEE Transactions on Multimedia, 2014, 16(5): 1282-1291. [PDF]

2. 王晗, 吴心筱*, 贾云得. 使用异构互联网图像组的视频标注. 计算机学报, 36(10), 2013. [PDF]

3. Yang Feng, Xinxiao Wu and Yunde Jia. Multi-group–multi-class domain adaptation for event recognition. IET Computer Vision, 2016.

4. Han Wang, Xinxiao Wu and Yunde Jia. Annotating videos from the web images. 21st International Conference on Pattern Recognition, 2012.