Skip to content

SemanticCollage: A semantically enriched digital mood board tool

Notifications You must be signed in to change notification settings

upq/SemanticCollage

 
 

Repository files navigation

SemanticCollage: Enriching Digital Mood Board Design with Semantic Labels

In the SemanticCollage research project (released July 6th 2020) we developed a digital mood board tool that attaches semantic labels to images by applying a state-of-the-art semantic labeling algorithm. It helps designers to 1) translate vague, visual ideas into search terms; 2) make better sense of and communicate their designs; while 3) not disrupting their creative flow.

Overview of the System:

SemanticCollage is a semantically enriched digital mood board tool for image collection and interpretation. Blue tools support image and text manipulation; Red tools provide semantic labels.

Getting started:

1) setup Postgres DB -- db name: semanticcollage
2) create tables using the statements in the DB folder
3) make sure the user 'research' has read/write privilages
(GRANT ALL PRIVILEGES ON DATABASE "semanticcollage" to research;)
4) install requirements: see requirements.tex
5) add a folder 'images' in the static folder

Usage:

1) Navigate to the folder.
2) start the system: python main.py
3) connect to: http://localhost:8080

Python version

2.7

Citation

@inproceedings{SemanticCollage,
  title={SemanticCollage: Enriching Digital Mood Board Design with Semantic Labels},
  author={Koch, Janin and Taffin, Nicolas and Lucero, Andr{\'e}s and Mackay, Wendy},
  booktitle={Proceedings of the Designing Interactive Systems Conference 2020 (DIS '20)},
  year={2020},
  organization={ACM}
}

License

This project is licensed under the MIT License, see the LICENSE.txt file for details.

Copyright © 2020 User Interfaces group, Aalto University, Finland

About

SemanticCollage: A semantically enriched digital mood board tool

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • JavaScript 79.0%
  • CSS 11.8%
  • HTML 7.4%
  • Python 1.8%