Art Recognition with a Computer Vision Model
Published Apr 20, 2022 • 1 min read

A semester's worth of art history is one API call away, thanks to the Wiki Art dataset and model on Roboflow Universe.

Computer Vision to Detect Art Styles

This classification model predicts what artistic movement and style an image falls within. The classes it can predict span a wide historical breadth from Early Renaissance to Pop Art.

Using the Art Detection Model

We trained the wiki art dataset using Roboflow's one-click training solution. The results are immediate and you can try it right now for free. On the wiki art project page drag & drop an image anywhere on the page to test model predictions.

Testing the art classification and detection model

This model is free to use, and we have a number of different deployment options available. You can infer via API, curl command, Webcam, or our Example Web App.

We also have sample code (here) you can copy and paste into your app.

Art Classification and Detection Dataset

This dataset is comprised of 6,417 images, with 25 different classes. You can find the original dataset by Wei Ren Tan, Chee Seng Chan, Hernan E. Aguirre, and Kiyoshi Tanaka on github, and their paper here.

New Datasets and Models Every Week

Keep an eye out on Roboflow Universe for new datasets and models you can use for free each week.

Cite this Post

Use the following entry to cite this post in your research:

Lukas Kelsey-Friedemann. (Apr 20, 2022). Art Recognition with a Computer Vision Model. Roboflow Blog: https://blog.roboflow.com/computer-vision-art-recognition/

Discuss this Post

If you have any questions about this blog post, start a discussion on the Roboflow Forum.

Written by

Lukas Kelsey-Friedemann