mirror of
https://github.com/openai/CLIP.git
synced 2026-11-05 01:37:26 +00:00
[GH-ISSUE #381] how to convert a picture to text embedding, without training these image model #326
Labels
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
openai/CLIP#326
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @yanchaoguo on GitHub (Aug 4, 2023).
Original GitHub issue: https://github.com/openai/CLIP/issues/381
Now, to generate specific concepts, like personal objects or artistic styles, by describing them using new "words" in the embedding space of pre-trained text-to-image models. These can be used in new sentences, just like any other word.
how to repace text embedding using a image without training
@wunting-chan commented on GitHub (Jan 20, 2024):
Not sure if you're looking for a solution without any training done on your end (meaning using models pretrained by others), or you're looking for a true no additional training solution.
If it's the former, you could look into ClipCap. They've trained a network to learn a prefix that gets added to an image's clip embedding for caption generation.
@LinMa910 commented on GitHub (Feb 21, 2025):
I also have the same question, how do you solvove this? Is there any presures that can share you solutions. very thank you for your help!