huggingface.co/Salesforce/blip-image-captioning-base
1 Users
0 Comments
22 Highlights
1 Notes
Tags
Top Highlights
vision-language tasks
bootstrapping the captions
the noisy ones
image-text retrieval
noisy image-text
Vision-Language Pre-training (VLP)
Bootstrapping Language-Image Pre-training
understanding-based tasks
generation-based tasks
excel in
a suboptimal source of supervision.
both vision-language understanding and generation tasks.
state-of-the-art results
a wide range of
synthetic captions
image captioning
average recall
VQA
transferred to
Glasp is a social web highlighter that people can highlight and organize quotes and thoughts from the web, and access other like-minded people’s learning.