metadata
license: cc-by-4.0
library_name: transformers
pipeline_tag: video-text-to-text
A competitive and human-aligned detailed video captioner model based on VILA-v1.5-13B and described in Cockatiel: Ensembling Synthetic and Human Preferenced Training for Detailed Video Caption.
This model produces detailed captions for input video, as presented in Cockatiel: Ensembling Synthetic and Human Preferenced Training for Detailed Video Caption.
For more details, please refer to our project page: https://sais-fuxi.github.io/projects/cockatiel