CVPR is the premier annual computer vision event comprising the main conference and several co-located workshops and short courses.

SkelNetOn 2019 is a Challenge for Geometric Shape Understanding workshop to utilize existing and develop novel deep learning architectures for shape understanding. SkelNetOn proposes three datasets and corresponding evaluation methodologies; all coherently bundled in three competitions with a dedicated workshop co-located with CVPR 2019 conference.

Parametric Track challenge is one of the challenges in SkelNetOn. In this challenge, the team of Dr. Liu Chang proposes an efficient and effective control point extraction algorithm for parametric skeleton generation. The object skeleton pixels are predicted via an hourglass network and partitioned into skeleton branches using Gaussian Mixture Models. For each skeleton branch, a Bezier curve is utilized to generate the control points. The radius of the skeleton is computed by the distance between the border of the object and the Bezier curve. The branches are sorted by the area so that the parametric skeleton representation is unique.

The team of Dr. Liu wins the first prize, in Parametric Track. The team includes Liu Chang, Wan Fang, Prof. Ye Qixiang from University of Chinese Academy of Sciences, Luo Dezhao, Zhang Yifei from Institute of Information Engineering，Chinese Academy of Sciences, Ke Wei from Carnegie Mellon University.

Congratulations to them!

CVPR全称IEEE Conference on Computer Vision and Pattern Recognition，是世界三大顶级计算机视觉会议（ICCV, CVPR和ECCV）之一，该会议包括主会议和几个位于同一地点的workshop以及short courses。

SkelNetOn是CVPR workshop中的一项竞赛，该竞赛利用现有的深度学习框架，开发新的形状理解结构来解决几何图形理解的问题。SkelNetOn 2019共有三个竞赛，并为三个竞赛提供了相应的数据集和评测方法。

Parametric Track是SkelNetOn的竞赛之一，在这项竞赛中，实验室刘畅博士的团队提出了一种高效的参数化骨架生成控制点提取算法。该算法通过沙漏网络对目标骨架像素进行预测，并利用高斯混合模型将其分割为骨架分支。对于每个骨架分支，使用贝塞尔曲线生成控制点。骨架的半径由对象边界和贝塞尔曲线之间的距离计算得出。分支按区域排序，以便参数化骨架表示是唯一的。

该团队获得了Parametric Track竞赛的第一名，团队成员包括来自中国科学院大学的刘畅、万方、叶齐祥（教授），中科院信工所罗德昭、张宜飞，卡内基梅隆大学的柯炜

祝贺他们！