| Summary: | With the growth of digital multimedia (especially images) in recent years, effective acquisition of proper metadata for images is still an open issue today. Crowd-based human computation represents a family of approaches able to provide large scale of metadata with decent quality. This thesis aims to provide a solution for a collaborative approach of image annotation. The solution proposed by this thesis uses a collaborative approach that includes the concepts of Games with a Purpose (GWAP) to enable identifying the more relevant areas of a photo. The idea behind using GWAP here, is to gather information based on user’s visual attention. The GWAP in this work depends on the ability of the user to quickly pay attention to parts of image instead of processing the whole scene of the image. This method helps the user to create tags on the image based on the important focus areas absorbed during the game. Thus, as a first objective, a game with a purpose was developed which motivates the user to identify regions of interest in the later part. The system developed includes the most important gamification elements expected to increase the user participation.
|
|---|