Skip to main content

Deep learning for text detection in the wild

Project Member(s): Jia, W.

Start year: 2015

Summary: Aims: This project aims to explore the state-of-the-art techniques of Deep Learning, which has recently become a hot topic in the filed of computer vision, and develop new deep features and deep learning techniques for fast and robust text detection in the wild. These new techniques play an important role in many computer vision applications and are expected to address the bottleneck problem of object detection and recognition research and applications. Objectives: The main objective is to undercover such new techniques and publish top quality papers on the selected topic. A second objective is to promote the Deep Learning techniques for other relevant computer vision, multimedia, and security domains. The above objectives align well with UTS Strategic Plan for 2009┬┐2018 which is to increase the scale, quality and impact of research in the University┬┐s discipline fields. Having top quality publications plays a significant role in succeeding research incomes from both governments and industry.

Publications:

Zhang, T, Jia, W, He, XS & Yang, J 2017, 'Discriminative Dictionary Learning with Motion Weber Local Descriptor for Violence Detection', IEEE Transactions on Circuits and Systems for Video Technology, vol. 27, no. 3, pp. 696-709.
View/Download from: UTS OPUS or Publisher's site

Zhang, T, Xu, L, Yang, J, Shi, P & Jia, W 2015, 'Sparse coding-based spatiotemporal saliency for action recognition', Proceedings - International Conference on Image Processing, ICIP, IEEE International Conference on Image Processing, IEEE, Quebec City, Canada, pp. 2045-2049.
View/Download from: UTS OPUS or Publisher's site

Keywords: Deep learning, text detection, robust

FOR Codes: Computer Vision, Image Processing, Expanding Knowledge in Technology