Video Prediction On Cityscapes 128X128
評価指標
Cond.
LPIPS
PSNR
Pred
SSIM
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | Cond. | LPIPS | PSNR | Pred | SSIM | Paper Title | Repository |
|---|---|---|---|---|---|---|---|
| SRVP | 10 | 0.447±0.014 | 20.97±0.43 | 20 | 0.603±0.016 | Stochastic Latent Residual Video Prediction | |
| Hier-VRNN | 2 | 0.264 ± .07 | - | 28 | 0.628±0.1 | Improved Conditional VRNNs for Video Prediction | |
| GHVAEs | 2 | 0.193 ± 0.014 | - | 28 | 0.740±0.4 | Greedy Hierarchical Variational Autoencoders for Large-Scale Video Prediction | - |
| SLAMP | 10 | 0.2941±0.022 | 21.73±0.76 | 20 | 0.649±0.025 | SLAMP: Stochastic Latent Appearance and Motion Prediction | |
| SVG (from Hier-VRNN) | 2 | 0.549 ± 0.06 | - | 28 | 0.574±0.08 | Stochastic Video Generation with a Learned Prior |
0 of 5 row(s) selected.