MILES: Visual BERT Pre-training with Injected Language Semantics for Video-text Retrieval

Benchmark Model Rank Results
zero-shot-video-retrieval-on-msvdMILES#9text-to-video R@1: 44.4text-to-video R@5: 76.2…
zero-shot-video-retrieval-on-lsmdcMILES#13text-to-video R@1: 11.1text-to-video R@5: 24.7…
zero-shot-video-retrieval-on-didemoMILES#18text-to-video R@1: 27.2text-to-video R@5: 50.3…
zero-shot-video-retrieval-on-msr-vttMILES#28text-to-video R@1: 26.1text-to-video R@5: 47.2…