2017-12-05
浏览次数:164This year , at the hands workshop of the International Conference on Computer Vision(ICCV, one of the top - level meetings of the computer vision) , held in Venice , Italy, guided graduate student Wu Wenbin, Li Chenyang and Cheng Zhuo, guided by Zhang Xin and Jin Lianwen, published paper “YOLSE: Egocentric Fingertip Detection from Single RGB Images” and successfully achieved the only Best Paper Award of the conference. An algorithm for detecting multiple gestures multiple fingertips in first angle projection of color videos is proposed in this paper, and achieved significant improvement in efficiency and accuracy. At the same time, the paper provided a detailly labeled hand gesture database for first angle projection(http://www.hcii-lab.net/data/), which contributes to the related research in the industry.
Hands workshop focuses on hand gesture recognition and hand attitude estimation. Every year, this conference brings together and shows the latest advances in hand estimation and recognition. The award for the best paper at the conference represents International experts’ great improvement for the work of the team of Zhang Xin and Jin Lianwen.
Jin Lianwen’s team aimed at frontier academic problems, and use deep learning and machine learning technology method for gesture recognition in human-computer interaction, text detection and recognition, face beautification and other fields has made many advanced research results. The paper has been included in the top-level conference and journal related fields. Warmly congratulate the teacher Jin Lianwen and Zhang Xin won ICCV Hands 2017 best paper award and hope the outstanding achievements in scientific research institute will achieve more success.