GIT: A Generative Image-to-text Transformer for Vision and Language
Paper
•
2205.14100
•
Published
•
1
GIT
GIT (Generative Image-to-text Transformer) is a model useful for vision-language tasks such as image/video captioning and question answering.
-
-
microsoft/git-base
Image-to-Text • Updated • 1.77M • 84 -
microsoft/git-large
Image-to-Text • Updated • 416 • 14 -
microsoft/git-base-vqav2
Visual Question Answering • Updated • 449 • 19 -
microsoft/git-large-coco
Image-to-Text • Updated • 6.95k • 101 -
microsoft/git-base-coco
Image-to-Text • Updated • 29.3k • 17 -
microsoft/git-large-r-textcaps
Image-to-Text • Updated • 158 • 10 -
microsoft/git-large-textcaps
Image-to-Text • Updated • 1.25k • 29 -
microsoft/git-base-vatex
Image-Text-to-Text • Updated • 503 • 5 -
microsoft/git-base-textcaps
Image-to-Text • Updated • 3.9k • 7 -
microsoft/git-large-r-coco
Image-to-Text • Updated • 132 • 10 -
microsoft/git-base-textvqa
Visual Question Answering • Updated • 1.74k • 6 -
microsoft/git-large-vqav2
Visual Question Answering • Updated • 132 • 17 -
microsoft/git-base-msrvtt-qa
Image-to-Text • Updated • 94 • 1 -
microsoft/git-large-vatex
Image-Text-to-Text • Updated • 335 • 1 -
microsoft/git-large-textvqa
Visual Question Answering • Updated • 72 • 4 -
microsoft/git-large-r
Image-to-Text • Updated • 62 • 2 -
microsoft/git-large-msrvtt-qa
Image-to-Text • Updated • 67 • 2