-
Notifications
You must be signed in to change notification settings - Fork 3.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Support for image search engines #317
base: main
Are you sure you want to change the base?
Conversation
@krasserm please read the following Contributor License Agreement(CLA). If you agree with the CLA, please reply with the following information.
Contributor License AgreementContribution License AgreementThis Contribution License Agreement (“Agreement”) is agreed to by the party signing below (“You”),
|
@microsoft-github-policy-service agree company="MerlinOne" |
This PR adds support for image search in the LAION-5B dataset (both, by image and text query) and makes a few minor refactorings. For example, when starting the demo UI with
python visual_chatgpt.py --load="ImageCaptioning_cuda:0,ImageEditing_cuda:1,Text2Image_cuda:2,ImageSearch_,VisualSearch_"
you can do something like:
Generate an image and find a visually similar image in the LAION-5B dataset. The generated image is used as query image:
Search an image in the LAION-5B dataset matching a user-provided description (query text) and then process that image with a visual foundation model:
Instruct Visual ChatGPT to generate an image, caption that image and then search for a LAION-5B image matching the caption. The instruction is decomposed by ChatGPT into three steps: image generation, image captioning and image search by query text:
These are rather simple examples. You can of course combine the search tools with other VFMs to support more complex image processing and search workflows.
Although image search engines are not VFMs I still hope this PR adds something useful to the project.