HIPI: Hadoop Image Processing Interface

The amount of images being uploaded to the internet is rapidly increasing, In most extensively used social media platform –Facebook users uploading over 2.5 billion new photos every month, however, applications that make use of this data are sternly lacking. Current computer vision applications use a small number of input images because of the difficulty is in attaining computational resources and storage options for huge amounts of data. As such, development of vision applications that use a large set of images has been inadequate. The Hadoop Mapreduce platform provides a system for large and computationally intensive distributed processing, though use of Hadoop system is rigorously limited by the technical complexities of developing useful applications. To ease this complexity, various image processing frameworks have been introduced which summaries the MapReduce model while converging on image processing task. Hadoop Image Processing Interface is one of those frameworks with many features and that objects to create an interface for computer vision with MapReduce technology.