Video Captioning On Vatex 1
Metrics
BLEU-4
CIDEr
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | BLEU-4 | CIDEr |
|---|---|---|
| vast-a-vision-audio-subtitle-text-omni-1 | 45.0 | 99.5 |
| nits-vc-system-for-vatex-video-captioning | 20.0 | 24.0 |
| icocap-improving-video-captioning-by | 36.9 | 63.4 |
| video-text-modeling-with-zero-shot-transfer | 39.7 | 77.8 |
| accurate-and-fast-compressed-video-captioning | 35.8 | 64.8 |
| diverse-video-captioning-by-adaptive-spatio | 36.25 | 65.07 |
| object-relational-graph-with-teacher | 32.1 | 49.7 |
| cosa-concatenated-sample-pretrained-vision | 43.7 | 96.5 |
| valor-vision-audio-language-omni-perception | 45.6 | 95.8 |
| icocap-improving-video-captioning-by | 37.4 | 67.8 |