PredRNN++: Towards A Resolution of the Deep-in-Time Dilemma in Spatiotemporal Predictive Learning

Benchmark Model Rank Results
video-prediction-on-kthPredRNN++#17SSIM: 0.865PSNR: 28.47Cond: 10Pred: 20
video-prediction-on-moving-mnistCausal LSTM#27MSE: 46.5MAE: 106.8SSIM: 0.898
video-prediction-on-synpickvpPredRNN++#2LPIPS: 0.053MSE: 51.73PSNR: 27.50SSIM: 0.894