It has been previously demonstrated that systems based on local features and relatively complex statistical models, namely 1D Hidden Markov Models (HMMs) and pseudo-2D HMMs, are suitable for face recognition. Recently, a simpler statistical model, namely the Gaussian Mixture Model (GMM), was also shown to perform well. In much of the literature devoted to these models, the experiments were performed with controlled images (manual face localization, controlled lighting, background, pose, etc.); however, a practical recognition system has to be robust to more challenging conditions. In this article we evaluate, on the relatively difficult BANCA database, the performance, robustness and complexity of GMM and HMM based approaches, using both manual and automatic face localization. We extend the GMM approach through the use of local features with embedded positional information, increasing performance without sacrificing its low complexity. Furthermore, we show that the traditionally used Maximum Likelihood (ML) training approach has problems estimating robust model parameters when there is only a few training images available; considerably more precise models can be obtained through the use of Maximum a Posteriori (MAP) training. We also show that face recognition techniques which obtain good performance on manually located faces do not necessarily obtain good performance on automatically located faces, indicating that recognition techniques must be designed from the ground up to handle imperfect localization. Finally, we show that while the pseudo-2D HMM approach has the best overall performance, authentication time on current hardware makes it impractical; the best trade-off in terms of authentication time, robustness and discrimination performance is achieved by the extended GMM approach. Optimal parameters for systems based on GMM (standard features), GMMext (extended features), 1D HMM and P2D HMM. ML: client models trained using traditional ML criterion;
We compare two classifier approaches, namely classifiers based on Multi Layer Perceptrons (MLPs) and Gaussian Mixture Models (GMMs), for use in a face verification system. The comparison is carried out in terms of performance, robustness and practicability. Apart from structural differences, the two approaches use different training criteria; the MLP approach uses a discriminative criterion, while the GMM approach uses a combination of Maximum Likelihood (ML) and Maximum a Posteriori (MAP) criteria. Experiments on the XM2VTS database show that for low resolution faces the MLP approach has slightly lower error rates than the GMM approach; however, the GMM approach easily outperforms the MLP approach for high resolution faces and is significantly more robust to imperfectly located faces. The experiments also show that the computational requirements of the GMM approach can be significantly smaller than the MLP approach at a cost of small loss of performance.
Ambient assisted living (AAL) has the ambitious goal of improving the quality of life and maintaining independence of older and vulnerable people through the use of technology. Most of the western world will see a very large increase in the number of older people within the next 50 years with limited resources to care for them. AAL is seen as a promising alternative to the current care models and consequently has attracted lots of attention. Recently, a number of researchers have developed solutions based on video cameras and computer vision systems with promising results. However, for the domain to reach maturity, several challenges need to be faced, including the development of systems that are robust in the real-world and are accepted by users, carers and society. In this literature review paper we present a comprehensive survey of the scope of the domain, the existing technical solutions and the challenges to be faced.
Lifestyle, or behavioural monitoring is an important element of telecare research where changes in activity profiles are used as a proxy to highlight a change in an individual"s health or care status. However, despite the promise of this approach for users and care providers it has been slow to develop. A literature review was undertaken to establish the current position with regard to lifestyle monitoring and to use this to inform requirements for the future development and implementation of such systems. In total, 74 papers met the inclusion criteria. Only 4 papers reported trials involving 20 or more individuals with a further 17 papers reporting trials involving one or more individuals. Most papers (n=53) were concerned with technology development initiatives. With respect to the technologies and strategies employed, motion monitoring dominated, followed by door and electrical appliance usage. The predominant monitoring strategy is that of detecting changes in activity levels. However, it was noticeable that little attention was given to determining when or how changes in the profile of activity should be used to raise a call for assistance from a health or care professional.
This paper details the results of a Face Authentication Test (FAT2004) [5] held in conjunction with the 17th International Conference on Pattern Recognition. The contest was held on the publicly available BANCA database [1] according to a defined protocol [7]. The competition also had a sequestered part in which institutions had to submit their algorithms for independent testing. 13 different verification algorithms from 10 institutions submitted results. Also, a standard set of face recognition software packages from the Internet [2] were used to provide a baseline performance measure.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.