rkouye/es-clip-image-search
Sample implementation of natural language image search with OpenAI's CLIP and Elasticsearch or Opensearch.
This project helps anyone with a large collection of images to search through them using everyday language, rather than relying on exact keywords or manual tags. You provide a collection of images and descriptions, and it gives you a web interface where you can type a phrase like "a dog playing fetch in a park" to find relevant pictures, even if they aren't explicitly tagged with those words. This is ideal for image library managers, content creators, or anyone needing to quickly retrieve specific visuals from a vast archive.
No commits in the last 6 months.
Use this if you need to build a user-friendly way for non-technical users to find images in a large, un-categorized or loosely categorized image database using natural language descriptions.
Not ideal if you only have a small number of images, or if your images are already perfectly tagged and categorized for traditional keyword search.
Stars
73
Forks
8
Language
Python
License
—
Category
Last pushed
Sep 01, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/rkouye/es-clip-image-search"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
Snehil-Shah/Multimodal-Image-Search-Engine
Text to Image & Reverse Image Search Engine built upon Vector Similarity Search utilizing CLIP...
comptech-winter-school/image-finder
Search relevant images by text query, using CLIP/ruCLIP embedder and HNSW index
R0GUE-A5H/CLIP-Style-Multimodal
A Multimodal AI Search Engine built from scratch using CLIP-style architecture (ViT + MPNet)....