Large-scale Video Interpolation Benchmark
LAVIB comprises a large collection of high-resolution videos sourced from the web. Metrics are computed for each video's motion magnitudes, luminance conditions, frame sharpness, and contrast. In total, LAVIB includes 283K clips from 17K ultra-HD videos, covering 77.6 hours. Benchmark train, val, and test sets maintain similar video metric distributions. Further splits are also created for out-of-distribution (OOD) challenges, with train and test splits including videos of dissimilar attributes.
Variants: LAVIB
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Video Frame Interpolation | FLAVR | LAVIB: A Large-scale Video Interpolation … | 2024-06-14 |
| Video Frame Interpolation | EMA-VFI | LAVIB: A Large-scale Video Interpolation … | 2024-06-14 |
| Video Frame Interpolation | RIFE | LAVIB: A Large-scale Video Interpolation … | 2024-06-14 |
Recent papers with results on this dataset: