EvalCrafter
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| AI-generated video detection | EvalCrafter | Floor33 Score99.77 | 42 | |
| Text-to-Video Generation | EvalCrafter | Text-Video Alignment75.62 | 34 | |
| AI-generated video detection | EvalCrafter 14 subsets (test) | Floor33 Score99.77 | 28 | |
| Video Generation | EvalCrafter | Visual Quality Score70.23 | 28 | |
| Text-to-Video Generation | EvalCrafter (test) | CLIPSIM0.191 | 21 | |
| Text-to-Video Generation | EvalCrafter Official (Four splits (Count, Color, Action, Others)) | T2V Alignment (Count)51.27 | 20 | |
| Video Detection | EvalCrafter | ACC96.01 | 14 | |
| Perceptual Video Quality Assessment | EvalCrafter | SRCC0.408 | 10 | |
| Text-to-Video Generation | EvalCrafter (full) | Visual Quality Score69.53 | 5 | |
| Auto-regressive scene extension | EvalCrafter | VQA_A75.48 | 5 | |
| Text-to-long video generation | EvalCrafter zero-shot 2048 videos, 64 frames | FVD85.83 | 5 | |
| Video-Text Prompt Alignment | EvalCrafter | SRCC0.586 | 4 |
Showing 12 of 12 rows