Dense Video Captioning On Vitt
Metrics
CIDEr
METEOR
SODA
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | CIDEr | METEOR | SODA |
|---|---|---|---|
| hicm-2-hierarchical-compact-memory-modeling | 51.2 | 9.6 | 0.150 |
| vid2seq-large-scale-pretraining-of-a-visual | 43.5 | 8.5 | 0.135 |
| Model 3 | 50.9 | 9.5 | 0.151 |
| end-to-end-dense-video-captioning-as-sequence-1 | 25.0 | 8.1 | - |