• Rupali Waghmode
• Prajakta Shinde
• Deepali Lokare
• Ashvini Nikhade

New approach has been given for selection of images which are appropriate for part of the visual summaries. The novel method present crowdsourcing and Infinite Push method for performing Visual Summarization of image collections. Most of visual summarization algorithms are guided by notions of relevance, representativeness and diversity of visual content. Explicit information on human’s perception regarding summarization process is not referred. This method is outlined on the base of how people generally thinking about generating summaries of image collections .Crowdsourcing experiment is performed for getting manually selected visual image summaries and method which help users for appropriate image selection. Based on insights of crowdsourcing experiment has purposed technique to select images automatically for generating visual summary. This technique utilizes the study various aspects of images like popularity of images, sentiment related to images, content and context of images. This technique defines images on the basis of their properties and semantic relation between images. It enables aesthetic features, sentiment analysis, and emotions associated with a specific group of images. Extraction of different features of images like luminance, colorfulness, sharpness, aspect ratio is done. These features are used to model image aesthetic appeal. The Infinite Push Ranking Method is used for training of crowsourced dataset. This system is excellent in terms of performance as compared with existing RankSVM method.

Crowdsourcing, Infinite Push Ranking, image aesthetic appeal

"Improved Algorithm for Visual Summarizarion of Image Collections using Crowdsourcing and Infinite Push Ranking", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 5, page no.273-277, May-2019, Available :http://www.jetir.org/papers/JETIRBO06052.pdf

"Improved Algorithm for Visual Summarizarion of Image Collections using Crowdsourcing and Infinite Push Ranking", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 5, page no. pp273-277, May-2019, Available at : http://www.jetir.org/papers/JETIRBO06052.pdf