Can LSTM model process Videos

[梅骁]

    I found that LSTM models in caffe only have constant time step. However,  the videos in dataset have different frames. Can caffe process this kind of sequential data? If It can, how can I do that. 

    Thank you! Looking forward your reply.

[Li Cheng]

I have the same question. Did you find any solution?

在 2016年12月10日星期六 UTC+8下午2:23:13，梅骁写道：