Pose Match is an interactive experience that matches your body pose to artworks from museum collections. Strike a pose in front of your webcam and discover artworks that share your stance.
This project uses machine learning (MoveNet) to detect body poses in real-time and compares them against a database of pre-analyzed artwork images.
1. Stand in front of your webcam
2. The system detects your pose
3. Your pose is matched to artworks with similar poses
4. Scan the QR code or click the image to learn more about the artwork
Explore mode will let you explore more deeply into the collection. Once you match with a work of art, that work will not repeat for some period of time.
Best Match mode will show you the best match to your pose. As a result, you are likely to see repeats of the same work more frequently.
Artwork images courtesy of The Metropolitan Museum of Art Open Access collection, the Smithsonian Open Access collection, and the SMK Open Access collection.
Built with ml5.js, p5.js, and TensorFlow by your friends at the GLAM-E Lab .