Liu Yuliang is a researcher and professor at the School of Artificial Intelligence and Automation, Huazhong University of Science and Technology. His research interests include artificial intelligence. He received his bachelor's and doctoral degrees from South China University of Technology, and during his doctoral studies, he conducted research as a visiting scholar at the University of Adelaide. He has also worked as a postdoctoral researcher at the University of Adelaide and the Chinese University of Hong Kong. He has led one project funded by the National Natural Science Foundation of China, and served as the principal investigator of two sub-projects of national key research and development programs, as well as own Excellent Doctoral Dissertation Award from CSIG. He has published over 10 first-author research papers in top-tier journals and conferences, such as IEEE TPAMI, IJCV, TIP, and CVPR. He has won 6 international competition championships and 2 Internet+ gold awards. He is a text aficionado, and always believes the advent of strong artificial intelligence.

Looking forward to students who are honest, upright, tolerant, optimistic, and have the awareness to think independently.

Partial works (First author)：

1. 【1】（CVPR Spotlight, ICDAR 2017 MLT Championship）Deep matching prior network: Toward tighter multi-oriented text detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 1962-1969), 2017.  

2. 【2】（Create CTW1500）Detecting curve text in the wild: New dataset and new solution. arXiv preprint arXiv:1712.02170, 2017. 

3. 【3】（IJCAI Oral, ICDAR 2019 ReCTS Championship）Omnidirectional scene text detection with sequential- free box discretization. International joint conferences on artificial intelligence organization, (pp. 3052-3058), 2019. 

4. 【4】（CVPR Oral, Full Strong Accepts）ABCNet: Real-time scene text spotting with adaptive bezier-curve network. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 9809-9818), 2020. 

5. 【5】（Pioneered Single-point Text Spotter）SPTS v2: Single-Point Scene Text Spotting. arXiv preprint arXiv:2301.01635, 2023. 

Source codes available at https://github.com/Yuliang-Liu.

Roadmap: Traditional Rectangle -> Tighter Quandrangle -> Compact Polygon -> Point is enough -> What's Next? (Upcoming：Relying on less annos through turning a clip model into a scene text spotter)