Michał Szachniewicz, Wojciech Kozłowski, Michał Stypułkowski, Maciej Zięba
TL;DR介绍了一种新型的自对抗方法 PointCAM，用于学习 3D 点云的遮罩函数，相较于之前的方法取得了更优秀的结果。
Abstract
self-supervised methods have been proven effective for learning deep representations of 3d point cloud data. Although recent methods in this domain often rely on random masking of inputs, the results of this appr