microsoft/art
Exploring the connections between artworks with deep "Visual Analogies"
Implements conditional image retrieval using deep semantic features from ImageNet-trained vision networks to enable specialized K-Nearest Neighbor queries across art collections. The system maps artworks into high-dimensional semantic space where distances are meaningful, allowing discovery of visual analogies across vastly different media, cultures, and time periods. Integrates collections from The Metropolitan Museum of Art and Rijksmuseum with a web interface for interactive exploration.
No commits in the last 6 months.
Stars
98
Forks
17
Language
TypeScript
License
MIT
Category
Last pushed
Jul 27, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/microsoft/art"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
kyegomez/ScreenAI
Implementation of the ScreenAI model from the paper: "A Vision-Language Model for UI and...
codetorex/spritex
A simple tool for extracting sprites from full frames. Useful for AI projects.
Zurdo1007/visual-intelligence
Visual Intelligence is a desktop app that extracts text from images and PDFs in Turkish and...
bluet/everypixel-js
JavaScript support for EveryPixel API
IntegerMan/AutomatingMyDog
An experiment in Azure Cognitive Services