Four Championships in Robust Reading International Competition

07 Sep 2015

On 28th August 2015, the outcomes of Robust Reading Competition were announced in the Thirteen IAPR International Conference on Document Analysis and Recognition (ICDAR), held in Nancy, France. Dr. Kaizhu Huang from Department of Electrical and Electronic Engineering teamed with Professor Xu-Cheng Yin from Beijing University of Science and Technology has won four championships in this premier international competition. Specifically, the team led by Dr. Huang and Prof. Yin were the winners in three challenged tasks: "Born-Digital End-to-End", "Focused Scene Text", and "Text in Video Localization/Tracking".

Robust Reading Competition was founded in 2003 by ICDAR. It has become the most important international competition in evaluating and verifying the latest research and technology in natural scenery image detection and recognition. Attracted by its great challenge and huge application potential, more than 100 research teams from various institutions, universities, and companies of tens of counties including US, China, Germany, France, Singapore, Russian, Japan, Korea, and India participated in the competition of 2015. Dr. Kaizhu Huang collaborated with Beijing University of Science and Technology achieved the best performance so far over the last 12 years. After the success of winning three championships in 2013, this is the second time that this research team took the first place and led the research at an international level.

Dr. Kaizhu Huang and his research team have made outstanding research achievements in machine learning and pattern recognition recently. Their research has been frequently published in the world-leading top journals such as IEEE Transactions on Pattern Analysis and Machine Intelligence, IEEE Transactions on Neural Networks and Learning Systems, and IEEE Transactions on Cybnetics. In addition, Dr. Huang has recently co-founded a new journal in Springer: BMC Big Data Analytics and serves as one of three key section editors in this journal.

International Conference on Document Analysis and Recognition has established itself as the premier conference in document analysis. This conference is held biennially by International Association of Pattern Recognition and it is one of major forums for researchers in pattern recognition and document analysis to exchange research ideas and new technologies.

07 Sep 2015