PexAce: A Method for Image Metadata Acquisition

Abstract

In this chapter, we present a semantics acquisition game for image tag acquisition, called PexAce. In this game, the player’s task is to (consecutively, in turns) disclose and conceal card pairs laid down on the game board, remember images on them and correctly identify identical image pairs. The game acquires the image descriptions through a game mechanics, which allows the players to make textual notes on what they see on the images. Acquired annotations are processed to tags by basic NLP and collaborative filtering. Experiments validating the game’s output with a general domain image dataset are presented. Furthermore, we present a modification of the game targeted on the acquisition of specific metadata for specific images: personal photography, which is otherwise a very complicated task for either automated or crowd-based approaches.