Skip to content

Latest commit

 

History

History
33 lines (16 loc) · 1012 Bytes

README.md

File metadata and controls

33 lines (16 loc) · 1012 Bytes

Human-Action-Recognition with ConvLSTM and LRCN

In this project, I have trained Convolution Long short term memory and Long-term Recurrent Convolutional Networks. The architecture are:

ConvLSTM:

conv

LRCN:

lrcn

I trained ConvLSTM for 25 epochs

Loss

ConvLstm Loss

Accuracy

ConvLstm Accuracy

I trained LRCN for 40 epochs

Loss

LRCN Loss

Accuracy

LRCN Accuracy