ClueMeIn: Obtaining More Specific Image Labels Through a Game

Christopher Harris


Abstract
The ESP Game (also known as the Google Image Labeler) demonstrated how the crowd could perform a task that is straightforward for humans but challenging for computers – providing labels for images. The game facilitated the task of basic image labeling; however, the labels generated were non-specific and limited the ability to distinguish similar images from one another, limiting its ability in search tasks, annotating images for the visually impaired, and training computer vision machine algorithms. In this paper, we describe ClueMeIn, an entertaining web-based game with a purpose that generates more detailed image labels than the ESP Game. We conduct experiments to generate specific image labels, show how the results can lead to improvements in the accuracy of image searches over image labels generated by the ESP Game when using the same public dataset.
Anthology ID:
2020.gamnlp-1.2
Volume:
Workshop on Games and Natural Language Processing
Month:
May
Year:
2020
Address:
Marseille, France
Editor:
Stephanie M. Lukin
Venue:
GAMESandNLP
SIG:
Publisher:
European Language Resources Association
Note:
Pages:
10–16
Language:
English
URL:
https://aclanthology.org/2020.gamnlp-1.2
DOI:
Bibkey:
Cite (ACL):
Christopher Harris. 2020. ClueMeIn: Obtaining More Specific Image Labels Through a Game. In Workshop on Games and Natural Language Processing, pages 10–16, Marseille, France. European Language Resources Association.
Cite (Informal):
ClueMeIn: Obtaining More Specific Image Labels Through a Game (Harris, GAMESandNLP 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.gamnlp-1.2.pdf