qq_38959366的博客

3DV: 3D Dynamic Voxel for Action Recognition in Depth Video(读书笔记)标题：3DV: 3D Dynamic Voxel for Action Recognition in Depth Video#CVPR 2020# 深度视频中的动作识别 3D Dynamic Voxel 方法作者： Yancheng Wang, Yang Xiao, Fu Xiong, Wenxiang Jiang, Zhiguo Cao, Joey Tianyi Zhou

（DPRL+GCNN读书笔记）Deep Progressive Reinforcement Learning for Skeleton-based Action Recognition1.摘要2.引言3.相关工作3.1Skeleton-based Action Recognition论文的特点：就是使用了DPRL方法，在全部的帧中挑选出了更重要的帧；使用更重要的帧进行识别。1.摘要 总结一下，首先作者在摘要中解释了贯穿全文的一个概念DPRL（Deep Progressive Reinforcement

Skeleton-Based Action Recognition with Directed Graph Neural Networks-DGNN读书笔记Skeleton-Based Action Recognition with Directed Graph Neural Networks1.摘要2. 引言3. 理论3.1骨骼信息3.2图构造3.3双向图神经网络（DGNN）3.3.1双向图神经网络 blockSkeleton-Based Action Recognition with Directed

Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition ------STGCN论文解读STGCN论文学习笔记1.摘要2.引言3.相关工作3.1 Neural Networks on Graphs3.2 Skeleton Based Action Recognition4.时空图卷积网络4.1bady parts4.2数据来源4.3骨架图的构造5.STGCN的原理新的甘特图功能，丰富你的文章UML