Skip to content
master
Switch branches/tags
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 
 
 
 
 
 
 
 
 
 
 

computer vision tests for facilitating non-text-based search and exploration

playing around with some computer vision and machine learning algorithms to hopefully reduce dependency on text-based searches of the DPLA corpus.

tests done in the “world war i posters” search in DPLA which contains about 2,300 images from different sources. a list of ids and urls can be found in data/images.ndjson

that query was chosen for the visual quality of the items (ideally posters, assuming the proper metadata exists in the records) and also relevant to 2018 being the 100 year anniversary of the end of world war i.

download the thumbnails

use the data/download.sh script to get all the thumbnails. the script requires jq

grid viewer demo

browse the posters in a grid viewer organized by visual similarity in a layout inspired by google images although this one makes use of css grid. works best on larger screens.

uses localForage for list management.

serendipity viewer demo

this viewer aims to allow serendipitous exploration of the posters. it chooses a random image from the set followed by other images from very similar to very dissimilar. it is somewhat inspired by the ios photos application, where the swipe-up gesture in an image displays related photos.

reverse image search web service

skeleton service that contains the trained model and can be used to find the similarity of any user-submitted image to the images in the set. more information is provided in that folder's README.

see also

this code borrows extensively from:

About

playing around with some computer vision and machine learning algorithms

Resources

License

Releases

No releases published

Packages

No packages published