توضیحات
Abstract
The amount of images being uploaded to the internet is rapidly in- creasing, with Facebook users uploading over 2.5 billion new pho- tos every month [Facebook 2010], however, applications that make use of this data are severely lacking. Current computer vision ap- plications use a small number of input images because of the dif- ficulty is in acquiring computational resources and storage options for large amounts of data [Guo. . . 2005; White et al. 2010]. As such, development of vision applications that use a large set of images has been limited [Ghemawat and Gobioff. . . 2003]. The Hadoop Mapreduce platform provides a system for large and com- putationally intensive distributed processing (Dean, 2004), though use of Hadoops system is severely limited by the technical com- plexities of developing useful applications [Ghemawat and Gob- ioff. . . 2003; White et al. 2010]. To immediately address this, we propose an open-source Hadoop Image Processing Interface (HIPI) that aims to create an interface for computer vision with MapRe- duce technology. HIPI abstracts the highly technical details of Hadoop’s system and is flexible enough to implement many tech- niques in current computer vision literature. This paper describes the HIPI framework, and describes two example applications that have been implemented with HIPI. The goal of HIPI is to create a tool that will make development of large-scale image processing and vision projects extremely accessible in hopes that it will em power researchers and students to create applications with ease-
INTRODUCTION
Many image processing and computer vision algorithms are appli-cable to large-scale data tasks. It is often desirable to run these algorithms on large data sets (e.g. larger than 1 TB) that are cur- rently limited by the computational power of one computer [GuoThese tasks are typically performed on a distributed system 2005].
by dividing the task across one or more of the following features: algorithm parameters, images, or pixels [White et al. 2010]. Per-forming tasks across a particular parameter is incredibly parallel and can often be perfectly parallel. Face detection and landmark classification are examples of such algorithms [Li and Crandall; Liu et al. 2009]. The ability to parallelize such tasks allows 2009 for scalable, efficient execution of resource-intensive applications
The MapReduce framework provides a platform for such applications-
Year : 2011
By : Chris Sweeney , Liu Liu , Sean Arietta , Jason Lawrence
File Information : English Language /5 Page /Size : 288 K
سال : 2011
کاری از : Chris Sweeney , Liu Liu , Sean Arietta , Jason Lawrence
اطلاعات فایل : زبان انگلیسی/ 5 صفحه / حجم : 288 K
نقد و بررسیها
هنوز بررسیای ثبت نشده است.