Multimedia Information Retrieval (MIR) at Leiden University

We live in the Age of Information. There is more information at our fingertips than at any other time in history. Portable digital devices such as mobile phones, streaming media services and social media are creating an explosion of multimedia. Producing and digitizing the massive amounts of multimedia information is straightforward and has resulted in petabytes of information just on the Internet.

The question we face is how to search and browse for the multimedia information that we are interested in? The faculty and students at Leiden University are engaged in this question and researching new paradigms and algorithms for finding multimedia information in the sea of data. To successfully do this we need to understand the content of the multimedia and the users.

Researchers and Collaborators

Dr. Erwin M. Bakker

Dr. Nies Huijsmans

Dr. Mark Huiskes (Shell)

Dr. Michael S. Lew

Dr. Ard Oerlemans (VDG)

Dr. Bart Thomee (Yahoo! Research)

Yanming Guo

Yu Liu

Song Wu

Student Research Projects at LIACS

Exploring in-game advertising in first person shooters

Content-based tag recommendation algorithms for unstructured data

Invariant Bag of Words for Image Retrieval

A Framework for Real-Time Face and Facial Feature Tracking using Optical Flow Pre-estimation and Template Tracking

Our Beautiful Planet

Sub-Image Search Engine


Improving SIFT accuracy with use of perspective transforms

GPU Based Generation and Real-Time Rendering of Semi-Procedural Terrain Using Features

Image Retrieval Using Local Feature Correspondences