search

microsoft / Oscar

Oscar and VinVL
MIT License
1.04k stars 251 forks source link

Which object labels (tags) did you use for NoCaps challenge (VinVL + VIVO model)? #187

Open enesmsahin opened 2 years ago

enesmsahin commented 2 years ago

Hi,

I am trying to reproduce your results for NoCaps challenge.

For the VinVL + VIVO model (NoCaps challenge) which object labels (tags) did you use for VIVO pretraining and COCO finetuning? Are tags from an object detector trained on Open Images dataset or from the same model which is used for feature extraction (trained on COCO + Objects365 + OpenImagesV5 + Visual Genome)?

Thanks.