Video Generation On Bair Robot Pushing
평가 지표
Cond
FVD score
Pred
Train
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | Cond | FVD score | Pred | Train |
|---|---|---|---|---|
| stochastic-video-generation-with-a-learned | 2 | 315.5 | 14 | 14 |
| latent-video-transformer | 1 | 320.9 | 15 | 15 |
| exploring-spatial-temporal-multi-frequency | 2 | 159.6 | 28 | 14 |
| stochastic-video-generation-with-a-learned | 2 | 256.62 | 28 | 10 |
| scaling-autoregressive-video-models | 1 | 94± 2 | 15 | 15 |
| stochastic-adversarial-video-prediction | 2 | 152±9 | 28 | 12 |
| nuwa-visual-synthesis-pre-training-for-neural | 1 | 86.9 | 15 | 15 |
| stochastic-adversarial-video-prediction | 2 | 143.43 | 28 | 10 |
| videoflow-a-flow-based-generative-model-for | 3 | 131±5 | 14 (total 16) | 10 |
| stochastic-variational-video-prediction | 2 | 262.5 | 14 | 14 |
| improved-conditional-vrnns-for-video | 2 | 149.22 | 28 | 10 |
| transformation-based-adversarial-video | 1 | 103.3 | 15 | 15 |
| stochastic-latent-residual-video-prediction-1 | 2 | 162 ± 4 | 28 | 12 |
| fitvid-overfitting-in-pixel-level-video | 1 | 93.6 | 15 | 15 |
| mocogan-decomposing-motion-and-content-for | 4 | 503 | 12 | 12 |
| efficient-video-generation-on-complex | 1 | 109.8 | 15 | 15 |
| stochastic-adversarial-video-prediction | 2 | 116.4 | 14 | 14 |
| diffusion-models-for-video-prediction-and | 1 | 84.20 | 15 | 20 |
| magvit-masked-generative-video-transformer | 1 | 62 | 15 | 15 |
| masked-conditional-video-diffusion-for | 2 | 87.9 | 14 | 5 |
| stochastic-video-generation-with-a-learned | 2 | 255±4 | 28 | 12 |
| stochastic-variational-video-prediction | 2 | 965±17 | 28 | 12 |
| improved-conditional-vrnns-for-video | 2 | 143.4 | 28 | 10 |
| stochastic-adversarial-video-prediction | 2 | - | 28 | 14 |
| masked-conditional-video-diffusion-for | 1 | 89.5 | 15 | 5 |
| latent-video-transformer | 1 | 125.76±2.90 | 15 | 15 |
| unsupervised-learning-for-physical | 2 | 296.5 | 14 | 14 |
| slamp-stochastic-latent-appearance-and-motion | 2 | 245 ± 5 | 28 | 10 |
| masked-conditional-video-diffusion-for | 2 | 118.4 | 28 | 5 |
| videogpt-video-generation-using-vq-vae-and | 1 | 103.3 | 15 | 15 |
| ccvs-context-aware-controllable-video | 1 | 99 ± 2 | 15 | 15 |