Video Summarization 벤치마크
Video Summarization on Shot2Story20K
CIDEr
- 2023-12-16 — SUM-shot: CIDEr 8.6
- 2024-05-31 — Shotluck-Holmes (3.1B): CIDEr 152.3
| Rank | Model | CIDEr | BLEU-4 | METEOR | ROUGE | Paper | Code | Year |
|---|---|---|---|---|---|---|---|---|
| 1 | Shotluck-Holmes (3.1B) | 152.3 | 7.67 | 23.2 | 43 | Shotluck Holmes: A Family of Efficient Small-Scale Large Language Vision Models For Video Captioning and Summarization | Skyline-9/Shotluck-Holmes | 2024 |
| 2 | SUM-shot | 8.6 | 11.7 | 19.7 | 26.8 | Shot2Story20K: A New Benchmark for Comprehensive Understanding of Multi-shot Videos | bytedance/Shot2Story | 2023 |
| 3 | Shotluck-Holmes (3.1B) | 152.3 | 7.67 | 23.2 | 43 | Shotluck Holmes: A Family of Efficient Small-Scale Large Language Vision Models For Video Captioning and Summarization | Skyline-9/Shotluck-Holmes | 2024 |
| 4 | SUM-shot | 8.6 | 11.7 | 19.7 | 26.8 | Shot2Story20K: A New Benchmark for Comprehensive Understanding of Multi-shot Videos | bytedance/Shot2Story | 2023 |