This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
|
||||||||
|
Paper Details
Paper Title
Survey on Use of Crowd sourced User Preferences for Visual Summarization of Image Collection
Authors
  Rupali Tanaji Waghmode,  Prajakta Subhash Shinde
Abstract
We present an algorithm for selecting suitable images for inclusion in visual summaries. This technique is designed on the basis of insights regarding how human perform visual summarization. For this purpose we use Amazon Mechanical Turk online Crowdsourcing Service. This algorithm use content and context of images, image sentiments, aesthetic features. We present a new solution based on text summarization and machine translation. We utilize pair wise ranking algorithm Rank.SVM for better performance. The studies conducted on a collection of geo-referenced Flickr image collections demonstrate the importance of our image selection approach.
Keywords- Crowdsourcing, aesthetic features, user-informed image selection
Publication Details
Unique Identification Number - IJEDR1801150Page Number(s) - 870-872Pubished in - Volume 6 | Issue 1 | March 2018DOI (Digital Object Identifier) -    Publisher - IJEDR (ISSN - 2321-9939)
Cite this Article
  Rupali Tanaji Waghmode,  Prajakta Subhash Shinde,   "Survey on Use of Crowd sourced User Preferences for Visual Summarization of Image Collection", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.6, Issue 1, pp.870-872, March 2018, Available at :http://www.ijedr.org/papers/IJEDR1801150.pdf
Article Preview
|
|
||||||
|