In this paper, we compare methods for estimating the complexity of two-dimensional shapes and introduce a method that exploits reconstruction loss of Variational Autoencoders with different sizes of latent vectors. Although complexity of a shape is not a well defined attribute, different aspects of it can be estimated. We demonstrate that our methods captures some aspects of shape complexity. Code and training details will be publicly available.
An Elastic Shape Variational Autoencoder for Skeleton Pose Trajectories
Systems and Information Engineering, University of Virginia(弗吉尼亚大学系统与信息工程系);Biomedical Engineering, Johns Hopkins University(约翰霍普金斯大学生物医学工程系);Dept. of Applied Mathematics and Statistics, Johns Hopkins University(约翰霍普金斯大学应用数学与统计学系)
Intraoperative Liver Surface Completion with Graph Convolutional VAE
Wellcome/EPSRC Centre for Interventional and Surgical Sciences, University College London(伦敦大学学院惠康/工程与物理科学研究委员会介入与外科科学中心);Centre For Medical Image Computing, University College London(伦敦大学学院医学图像计算中心);Division of Surgery and Interventional Science, University College London(伦敦大学学院外科与介入科学系)
On the effectiveness of GAN generated cardiac MRIs for segmentation
Université de Bourgogne Franche-Comté(勃艮第-弗朗什-孔泰大学);Cardiac Simulation & Imaging Software(心脏模拟与成像软件公司);University of Sherbrooke(舍布鲁克大学);University Hospital of Dijon(第戎大学医院)