alexlee-gk / video_prediction

Stochastic Adversarial Video Prediction
https://alexlee-gk.github.io/video_prediction/
MIT License
303 stars 65 forks source link

CDNA Masks #48

Open LukeDitria opened 2 years ago

LukeDitria commented 2 years ago

Just a question with the number of Masks for CDNA in the sv2p_model. You have a mask for each of the 10 Transformed outputs, another for a static background (masking the input image) and another for the output of the model, should that not be 12 masks, not 11? It seems as though you are not using one of the transformed outputs, or am I interpreting this wrong? Thanks