search

microsoft / UniVL

An official implementation for " UniVL: A Unified Video and Language Pre-Training Model for Multimodal Understanding and Generation"
https://arxiv.org/abs/2002.06353
MIT License
339 stars 54 forks source link

When will you release the code of 'Action Step Localization' and 'Action Segmentation' tasks? #3

Closed butterluo closed 3 years ago

butterluo commented 3 years ago

When will you release the code of 'Action Step Localization' and 'Action Segmentation' tasks?

ArrowLuo commented 3 years ago

We have no plan to release the 'Action Step Localization,' 'Action Segmentation,' and the 'Multimodal Sentiment Analysis' recently due to the extra clean job and checks. And may not be open source those parts in the future. If you have any questions on the details of implementation ignored in our paper, you can write to us via email or this issue.