##SMEM This repository has the source code for the paper "An Improved LSTM-based Network: Learning Explicit Shape and Motion Evolution Maps for Skeleton-based Human Action Revognition"

##Dependencies

• Theano
• Keras
• Scipy
• matplotlib
• Numpy

##Source Code Description

We implement our network based on Keras. Keras supports custom operation, so several novel layers proposed in our paper can be easily implemented. Also, the LSTM-based architecture of our method can be easily implemented with Keras Functional API. Some files are described as follows.

• net.py provides the code for overall fusion model (SMEM)
• data_shape.py provides the code for shape evolution maps (SEM)
• data_motion.py provides the code for motion evolution maps (MEM)
• main.py provides the code for main exe file

others: kutilities : provides the code for weighted aggregate layer (WAL), needing compile the setup.py to setup mul: provides the fusion model (SMEM) mul_WAL: provides the fusion model with WAL (SMEM + WAL)

about how to obtain the SEM and MEM, you can easily implement it.

our experimental NTU RGB+D dataset's skeleton data, you can download at: http://rose1.ntu.edu.sg/datasets/actionrecognition.asp