“If I could tell the story in words, I wouldn’t need to lug a camera.” -Lewis Hine
The word “photograph” and our understanding of it has evolved from an upside down image with color and preserved perspective to sharing photos on the phone. In today’s world photograph has gained a whole different meaning. It is now possible to share, view, edit, or apply filters on photos that we take. According to an estimate, a total of 300 million photos are uploaded on Facebook every day.
Many solutions to research questions have been found using digital data objects that form a brilliant scenario along with their meta-data and social ecosystem. Media collections are valuable for practical research in many domains. Scientific researches in the fields like computer vision, social computing and artificial intelligence heavily rely on collected media data sets which makes research complex and complicated.
Experimentation, validation, analysis of designs and building applications, it all requires data. However, a big problem is that data sets are not often made available publicly. Moreover, they only contain a specific kind of media (photos, videos or audios), rather than a mixture of them. This is where the creation of YFCC100M (Yahoo Flickr Creative Commons 100 Million Data set) comes into play. With a grand total of `100 million data objects (99.2 million photos and 0.8 million videos, uploaded between (2004 to 2014), YFCC100M is the largest multimedia collection ever released to public. It is completely free and legal to use, promoting existing and future experiments.
Find out more in the video as Bart Thomee of Yahoo Labs discusses about YFCC100M.
Reference Link: https://zephoria.com/top-15-valuable-facebook-statistics/