3huo / 3DV-Action

Action Recognition CVPR2020

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

3DV: 3D Dynamic Voxel for Action recognition in depth video.

Paper pdf

To be represented in CVPR2020

Usege:

Our model consists of 2 parts. The first part is about 3DV point data generation and apprearance point data preparation. For example, "3DV_construction\python\ntu120_3dv_pre.py" is used to generate the 3DV point data with multiple motion features. The output 3DV points data form is N*(x,y,z,m_g,m_1,...,m_T) "3DV_construction\python\RawDepthPoints_appearance_pre.py" is used to sample multi-segment temporal frames with its raw points into the form of N*(x,y,z).

The second part is about multi-stream PointNet++. The generated point data in part 1 can be fed to Network.

Some related data:

https://pan.baidu.com/s/1TtApXBQx4sNi4ZmTOL6lSw code:k0jt

About

Action Recognition CVPR2020


Languages

Language:Python 80.0%Language:MATLAB 20.0%