Pytorch implementation of SpatioTemporal AutoEncoder
Paper: Abnormal Event Detection in Videos using Spatiotemporal Autoencoder(https://arxiv.org/pdf/1701.01546.pdf)
I tried to implement the ST_AE with Pytorch, referring to keras code from https://github.com/aninair1905/Abnormal-Event-Detection. There could be incorrect implementation in my code, so feel free to announce me about error.
In referred Keras's code(https://github.com/aninair1905/Abnormal-Event-Detection), there are dropout and recurrent_drop out in ConvLSTM block. I cannot add them to my ConvLSTM.py.