VLAB: Enhancing Video Language Pre-training by Feature Adapting and Blending

Benchmark Model Rank Results
video-captioning-on-msr-vttVLAB–CIDEr: 74.9METEOR: 33.4ROUGE-L: 68.3BLEU-4: 54.6
video-captioning-on-msvdVLAB–CIDEr: 179.8BLEU-4: 79.3METEOR: 51.2ROUGE-L: 87.9
video-retrieval-on-didemoVLAB–text-to-video R@1: 56.8text-to-video R@5: 81.6…
video-retrieval-on-msr-vttVLAB–text-to-video R@1: 55.1text-to-video R@5: 78.8…
video-retrieval-on-msvdVLAB–text-to-video R@1: 57.5text-to-video R@5: 83.6…
visual-question-answering-on-msrvtt-qaVLAB–Accuracy: 0.496
visual-question-answering-on-msvd-qaVLAB–Accuracy: 0.61