Multimodal Search

Embedding-based image and text retrieval for a folder of images. Supports image/text/patch search. Useful for exploring and viewing your images.

Getting Started

Move your folder of images to ./images/. Run scripts/clip_embed.py and scripts/dino_embed.py on the images to produce vectors.

Launch the app with docker compose up. Open http://localhost:3000 to see the UI.

Name		Name	Last commit message	Last commit date
Latest commit History 8 Commits
assets		assets
scripts		scripts
static		static
.gitignore		.gitignore
.python-version		.python-version
Dockerfile		Dockerfile
LICENSE		LICENSE
README.md		README.md
docker-compose.yml		docker-compose.yml
embedder.py		embedder.py
knn.py		knn.py
nginx.conf		nginx.conf
pyproject.toml		pyproject.toml
uv.lock		uv.lock