search

LiuRicky / ts2_net

[ECCV2022] A pytorch implementation for TS2-Net: Token Shift and Selection Transformer for Text-Video Retrieval
75 stars 9 forks source link

About ActivityNet-Caption, DiDeMo, LSMDC experiment's Detail #9

Closed sssss99999 closed 1 year ago

sssss99999 commented 1 year ago

First of all, thank you for sharing your good research. There are only MSRVTT and VATEX datasets training parameters in scripts, and ActivityNet, DiDeMo, and LSMDC are not available. Can you provide them?

LiuRicky commented 1 year ago

First of all, thank you for sharing your good research. There are only MSRVTT and VATEX datasets training parameters in scripts, and ActivityNet, DiDeMo, and LSMDC are not available. Can you provide them?

Thanks for your attention to our work.

The scripts of datasets you mentioned is the same as msrvtt and vatex. You can replace '--data_path ${DATA_PATH}/vatex_data --features_path ${DATA_PATH}/compressed_videos --output_dir ckpts/vatex --datatype vatex ' with the dataset you want to use. Besides, if your data annotation is not suitable with specific dataloader, you may need to change dataloader_xxxx_retrieval.py