At ICMR 2014, we will be running two tutorials, these are as follows:
Visual recognition in large collections (Hervé Jégou)
The tutorial covers state-of-the-art methods and systems for visual recognition in large collections of images, considering different trade-offs with respect to efficiency and search quality.
The tutorial first focuses on three important ingredients of such systems: 1) The image description itself, including the bag-of-words model, Fisher vectors, and more generally match kernels; 2) Algorithms for efficient similarity search, in particular approximate nearest neighbour search, compressed-domain search and Min-hash; 3) Complementary techniques, aiming in particular at exploiting the spatial information on a large scale and query expansion.
Then, the tutorial gives a brief overview of the problem of automatic discovery of visual patterns or related images in unstructured image collections. The tutorial finally describes a recent approach for precise search in very large collections of videos with limited resources.
Music Information Retrieval - Theory and Applications (George Tzanetakis)
The goal of this turorial is to provide a thorough theoretical overview of the state-of-the-art in Music Information Retrieval followed by a practical hands-on demonstration of several existing tools and resources that can be used for research in this area.
Specic emphasis will be given on how MIR techniques relate to other elds of current multimedia research. MIR is an inherently interdisciplinary area touching on several research areas such as digital signal processing, machine learning, perception, visualization, human computer interaction, content-based retrieval and digital libraries. Music has several unique characteristics that differentiate it from other areas of multimedia research. The different problems and techniques proposed to solve them that will described in the rst part of the tutorial will be followed with concrete practical examples of applying these techniques using existing software tools and datasets.