Video Synthesis on Urban Street scenario n=7
1.392Inference Time (s)X-Cache
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| X-CacheCamera=F-C (front centre)2026.04 | 1.392 | 53.83 | 99.88 | 0.0004 | 71.4 | 2.65 | |
| X-CacheCamera=F-W (front wide)2026.04 | 1.392 | 50.27 | 99.87 | 0.0004 | 71.4 | 2.65 | |
| X-CacheCamera=S-FL2026.04 | 1.392 | 49.49 | 99.85 | 0.0005 | 71.4 | 2.65 | |
| X-CacheCamera=S-FR2026.04 | 1.392 | 48.69 | 99.84 | 0.0005 | 71.4 | 2.65 | |
| X-CacheCamera=S-RL2026.04 | 1.392 | 48.59 | 99.85 | 0.0005 | 71.4 | 2.65 | |
| X-CacheCamera=S-RR2026.04 | 1.392 | 48.07 | 99.85 | 0.0005 | 71.4 | 2.65 | |
| X-CacheCamera=Rear2026.04 | 1.392 | 51.77 | 99.86 | 0.0005 | 71.4 | 2.65 | |
| X-CacheCamera=7-cam (shaded)2026.04 | 1.392 | 51.37 | 99.9 | 0.0003 | 71.4 | 2.65 | |
| Baseline full-compute reference2026.04 | 3.682 | — | — | — | — | — |