SwinLSTM: Improving Spatiotemporal Prediction Accuracy using Swin Transformer and LSTM

Benchmark Model Rank Results
video-prediction-on-human36mSwinLSTM#2SSIM: 0.913MSE: 332MAE: 1190
video-prediction-on-moving-mnistSwinLSTM#8MSE: 17.7SSIM: 0.962