Indexing of Large Data Sets in Image Retrieval

Visual data, especially in the form of images and videos, is created in abundance nowadays. However, search in big amounts of visual data are still a challenging topic. Everything that fits into RAM is easy to handle, but large scale data sets need to distributed or held on slow storage devices, and big data is too large even for single devices and small networks. While image search and classification has arrived at a stage, where users are impressed by the accuracy of search results in many cases, indexing and retrieval from the index are hot topics. In this talk I will present the most common practical methods for indexing and retrieval visual information and I will point out challenges and opportunities in this field.