Video Synthesis on Cataract-1k (test)
73FVDSG2VID
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| SG2VIDConditioning=Scene graphs2026.05 | 73 | 14.9 | 0.392 | 60.7 | 62.3 | |
| SG2VID2025.06 | 77 | 15.5 | 0.397 | — | — | |
| SWoMoConditioning=Full (Scene graphs + Simulated sequence)2026.05 | 123 | 20.1 | 0.388 | 64.5 | 65.6 | |
| SWoMo-✗IMGAblation=No image conditioning2026.05 | 134.3 | 20.2 | 0.389 | 62.2 | 64.2 | |
| EndoraGeneration Mode=Unconditional2026.05 | 258.7 | 40 | 0.379 | — | — | |
| Endora2025.06 | 265.9 | 30.2 | 0.377 | — | — | |
| SG2VID-XIMG2025.06 | 278.4 | 33.3 | 0.409 | — | — | |
| SWoMo-✗SGAblation=No scene graph conditioning2026.05 | 283.6 | 35.5 | 0.434 | 52.9 | 55.4 | |
| StyleGAN-V2025.06 | 442.6 | 118.2 | 0.286 | — | — | |
| StyleGAN-VGeneration Mode=Unconditional2026.05 | 501.2 | 116.4 | 0.286 | — | — | |
| MOFAConditioning=Initial frame and motion trajectory2026.05 | 716.7 | 94.6 | 0.358 | 41.8 | 40.4 | |
| MOFA2025.06 | 722.2 | 89.9 | 0.361 | — | — | |
| MedSoraGeneration Mode=Unconditional2026.05 | 809.9 | 147.6 | 0.324 | — | — | |
| MedSora2025.06 | 901.8 | 137.6 | 0.324 | — | — | |
| LVDMConditioning=Text-based scene graph triplets2026.05 | 1,469.6 | 176.6 | 0.519 | 21.3 | 18.8 | |
| LVDM2025.06 | 1,656.6 | 186 | 0.534 | — | — |