MDMMT-2: Multidomain Multimodal Transformer for Video Retrieval, One More Step Towards Generalization

Benchmark Model Rank Results
video-retrieval-on-lsmdcMDMMT-2text-to-video R@1: 26.9text-to-video R@5: 46.7
video-retrieval-on-msr-vttMDMMT-2text-to-video R@1: 33.7text-to-video R@5: 60.5
video-retrieval-on-msvdMDMMT-2text-to-video R@1: 56.8text-to-video R@5: 83.1
video-retrieval-on-youcook2MDMMT-2text-to-video R@1: 32.0text-to-video R@5: 64.0