26hzhang / ReLoCLNet

Video Corpus Moment Retrieval with Contrastive Learning (SIGIR 2021)

26hzhang/ReLoCLNet Issues

some question about activitynet dataset
Updated a month ago1
some questions about model's performence on activitynet
Updated a month ago1
in moment localization module, why u use simarilty to predict start/end prob? is this design due to frame level contrastive learning?
Updated 2 years ago
Cant download the dataset feature.
Updated 2 years ago
Pre-trained model
Updated 3 years ago