This paper describes the Institute for Infocomm Research (IIR) system for the 2007 Language Recognition Evaluation (LRE) conducted by the National Institute of Standards and Technology (NIST). The submitted system is a fusion of multiple state-ofthe-art language classifiers using diversified discriminative language cues. We implemented several state-of-the-art algorithms using both phonotactic ...

This paper deals with the concept of correlation in multi-classifier fusion tasks. The need to encapsulate the degree of correlation stems from the notion that the least amount of correlation among the classifiers is desired in fusion tasks. We define a measure that captures the correlation for n classifiers for binary output. In addition, we propose correlation calculation for classifiers with...

This paper discusses the use of a secondary likelihood classifier scheme for improving speaker recognition performance. The system models the output likelihoods of a typical Gaussian Mixture Model system across multiple speakers. The Output Probability Distributions (OPD) of the primary classifiers contain information on inter-speaker relationships, and are modelled by secondary classifiers to ...

A generic fusion problem is studied for multiple sensors whose outputs are probabilistically related to their inputs according to unknown distributions. Sensor measurements are provided as iid input-output samples, and an empirical risk minimization method is described for designing fusers with distribution-free performance bounds. The special cases of isolation and projective fusers for classi...

This paper deals with the concept of correlation in multi-classifier fusion tasks. The need to encapsulate the degree of correlation stems from the notion that the least amount of correlation among the classifiers is desired in fusion tasks. We define a measure that captures the correlation for n classifiers for binary output. In addition, we propose correlation calculation for classifiers with...

Multiple people detection and tracking is a challenging task in real-world crowded scenes. In this paper, we have presented an online multiple people tracking-by-detection approach with a single camera. We have detected objects with deformable part models and a visual background extractor. In the tracking phase we have used a combination of support vector machine (SVM) person-specific classifie...

As mobile devices, intelligent displays, and home entertainment systems permeate digital markets, the desire for users to interact through spoken and visual modalities similarly grows. Previous interactive systems limit voice activity detection (VAD) to the acoustic domain alone, but the incorporation of visual features has shown great improvement in performance accuracy. When employing both ac...

After deploying a classifier in production it is essential to support its lifecycle. This paper describes the application of an ensemble of classifiers to support two stages of the lifecycle of an on-line classifier used to underwrite life insurance applications: the monitoring of its decisions quality and the updating of the production classifier over time. All combinations of five classificat...

Multimedia Event Detection is a multimedia retrieval task with the goal of finding videos of a particular event in an internet video archive, given example videos and descriptions. We focus here on mining features of example videos to learn the most characteristic features, which requires a combination of multiple complementary types of features. Generally, early fusion and late fusion are two ...