Sense-GVT / DeCLIP

Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigm
622 stars 31 forks source link

Could you provide model weights of the pretrained text encoder? #2

Closed RenShuhuai-Andy closed 2 years ago

zlccccc commented 2 years ago

Released.