Face recognition using fractal codes

In this paper we propose a new method for face recognition using fractal codes. Fractal codes represent local contractive, affine transformations which when iteratively applied to range-domain pairs in an arbitrary initial image result in a fixed point close to a given image. The transformation parameters such as brightness offset, contrast factor, orientation and the address of the corresponding domain for each range are used directly as features in our method. Features of an unknown face image are compared with those pre-computed for images in a database. There is no need to iterate, use fractal neighbor distances or fractal dimensions for comparison in the proposed method. This method is robust to scale change, frame size change and rotations as well as to some noise, facial expressions and blur distortion in the image

Face authentication test on the BANCA database

This work details the results of a face authentication test (FAT2004) (http://www.ee.surrey.ac.uk/banca/icpr2004) held in conjunction with the 17th International Conference on Pattern Recognition. The contest was held on the publicly available BANCA database (http://www.ee.surrey.ac.uk/banca) according to a defined protocol (E. Bailly-Bailliere et al., June 2003). 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 (http://www.cs.colostate.edu/evalfacerec) were used to provide a baseline performance measure.

Super-resolved faces for improved face recognition from surveillance video advances in biometrics

Characteristics of surveillance video generally include low resolution and poor quality due to environmental, storage and processing limitations. It is extremely difficult for computers and human operators to identify individuals from these videos. To overcome this problem, super-resolution can be used in conjunction with an automated face recognition system to enhance the spatial resolution of video frames containing the subject and narrow down the number of manual verifications performed by the human operator by presenting a list of most likely candidates from the database. As the super-resolution reconstruction process is ill-posed, visual artifacts are often generated as a result. These artifacts can be visually distracting to humans and/or affect machine recognition algorithms. While it is intuitive that higher resolution should lead to improved recognition accuracy, the effects of super-resolution and such artifacts on face recognition performance have not been systematically studied. This paper aims to address this gap while illustrating that super-resolution allows more accurate identification of individuals from low-resolution surveillance footage. The proposed optical flow-based super-resolution method is benchmarked against Baker et al.’s hallucination and Schultz et al.’s super-resolution techniques on images from the Terrascope and XM2VTS databases. Ground truth and interpolated images were also tested to provide a baseline for comparison. Results show that a suitable super-resolution system can improve the discriminability of surveillance video and enhance face recognition accuracy. The experiments also show that Schultz et al.’s method fails when dealing surveillance footage due to its assumption of rigid objects in the scene. The hallucination and optical flow-based methods performed comparably, with the optical flow-based method producing less visually distracting artifacts that interfered with human recognition.

Least-squares congealing for large numbers of images

In this paper we pursue the task of aligning an ensemble of images in an unsupervised manner. This task has been commonly referred to as “congealing” in literature. A form of congealing, using a least-squares criteria, has been recently demonstrated to have desirable properties over conventional congealing. Least-squares congealing can be viewed as an extension of the Lucas & Kanade (LK)image alignment algorithm. It is well understood that the alignment performance for the LK algorithm, when aligning a single image with another, is theoretically and empirically equivalent for additive and compositional warps. In this paper we: (i) demonstrate that this equivalence does not hold for the extended case of congealing, (ii) characterize the inherent drawbacks associated with least-squares congealing when dealing with large numbers of images, and (iii) propose a novel method for circumventing these limitations through the application of an inverse-compositional strategy that maintains the attractive properties of the original method while being able to handle very large numbers of images.

Make-buy decisions in the face of technological change: does industry clockspeed matter?

Some evidence in the area of make-buy decisions for new technologies suggests that it is a good idea for a company to pursue a fairly rigorous ''make'' policy in the early days of a potentially disruptive innovation. Other studies prescribe exactly the opposite, promoting instead a ''buy'' strategy. This paper seeks to bridge the gap between these perspectives by suggesting that both strategies are valid, but that they are most successfully applied in different market environments. The ''make'' prescription may be more suited to either extremely fast or extremely slow rates of technological change, while a ''buy'' strategy might be more appropriate in market sectors where technologies evolve at a medium pace. This paper highlights the importance of industry clockspeed and supplier relationships in make-buy decisions for new technologies, and puts forward two new hypotheses that require empirical testing.

Reshaping Doctoral Education : International Approaches and Pedagogies

The number of doctorates being awarded around the world has almost doubled over the last ten years, propelling it from a small elite enterprise into a large and ever growing international market. Within the context of increasing numbers of doctoral students this book examines the new doctorate environment and the challenges it is starting to face. Drawing on research from around the world the individual authors contribute to a previously under-represented focus of theorising the emerging practices of doctoral education and the shape of change in this arena. Key aspects, expertly discussed by contributors from the UK, USA, Australia, New Zealand, China, South Africa, Sweden and Denmark include: -the changing nature of doctoral education -the need for systematic and principled accounts of doctoral pedagogies -the importance of disciplinary specificity -the relationship between pedagogy and knowledge generation -issues of transdisciplinarity. Reshaping Doctoral Education provides rich accounts of traditional and more innovative pedagogical practices within a range of doctoral systems in different disciplines, professional fields and geographical locations, providing the reader with a trustworthy and scholarly platform from which to design the doctoral experience. It will prove an essential resource for anyone involved in doctorate studies, whether as students, supervisors, researchers, administrators, teachers or mentors.

Face recognition across pose on video using eigen light-fields

We propose an approach to employ eigen light-fields for face recognition across pose on video. Faces of a subject are collected from video frames and combined based on the pose to obtain a set of probe light-fields. These probe data are then projected to the principal subspace of the eigen light-fields within which the classification takes place. We modify the original light-field projection and found that it is more robust in the proposed system. Evaluation on VidTIMIT dataset has demonstrated that the eigen light-fields method is able to take advantage of multiple observations contained in the video.

Helping undergraduate students across disciplines and cultures actively engage and collaborate as equal members of a community of scholars : peer review within an e-learning environment

At the international level, the higher education sector is currently being subjected to increased calls for public accountability and the current move by the OECD to rank universities based on the quality of their teaching and learning outcomes. At the national level, Australian universities and their teaching staff face numerous challenges including financial restrictions, increasing student numbers and the reality of an increasingly diverse student population. The Australian higher education response to these competing policy and accreditation demands focuses on precise explicit systems and procedures which are inflexible and conservative and which ignore the fact that assessment is the single biggest influence on how students approach their learning. By seriously neglecting the quality of student learning outcomes, assessment tasks are often failing to engage students or reflect the tasks students will face in the world of practice. Innovative assessment design, which includes new paradigms of student engagement and learning and pedagogically based technologies have the capacity to provide some measure of relief from these internal and external tensions by significantly enhancing the learning experience for an increasingly time-poor population of students. That is, the assessment process has the ability to deliver program objectives and active learning through a knowledge transfer process which increases student participation and engagement. This social constructivist view highlights the importance of peer review in assisting students to participate and collaborate as equal members of a community of scholars with both their peers and academic staff members. As a result of increasing the student’s desire to learn, peer review leads to more confident, independent and reflective learners who also become more skilled at making independent judgements of their own and others' work. Within this context, in Case Study One of this project, a summative, peer-assessed, weekly, assessment task was introduced in the first “serious” accounting subject offered as part of an undergraduate degree. The positive outcomes achieved included: student failure rates declined 15%; tutorial participation increased fourfold; tutorial engagement increased six-fold; and there was a 100% student-based approval rating for the retention of the assessment task. However, in stark contrast to the positive student response, staff issues related to the loss of research time associated with the administration of the peer-review process threatened its survival. This paper contributes to the core conference topics of new trends and experiences in undergraduate assessment education and in terms of innovative, on-line, learning and teaching practices, by elaborating the Case Study Two “solution” generated to this dilemma. At the heart of the resolution is an e-Learning, peer-review process conducted in conjunction with the University of Melbourne which seeks to both create a virtual sense of belonging and to efficiently and effectively meet academic learning objectives with minimum staff involvement. In outlining the significant level of success achieved, student-based qualitative and quantitative data will be highlighted along with staff views in a comparative analysis of the advantages and disadvantages to both students and staff of the staff-led, peer review process versus its on-line counterpart.

Developing a Second Life virtual field trip for university students : an action research approach

Background: Integrating 3D virtual world technologies into educational subjects continues to draw the attention of educators and researchers alike. The focus of this study is the use of a virtual world, Second Life, in higher education teaching. In particular, it explores the potential of using a virtual world experience as a learning component situated within a curriculum delivered predominantly through face-to-face teaching methods. Purpose: This paper reports on a research study into the development of a virtual world learning experience designed for marketing students taking a Digital Promotions course. The experience was a field trip into Second Life to allow students to investigate how business branding practices were used for product promotion in this virtual world environment. The paper discusses the issues involved in developing and refining the virtual course component over four semesters. Methods: The study used a pedagogical action research approach, with iterative cycles of development, intervention and evaluation over four semesters. The data analysed were quantitative and qualitative student feedback collected after each field trip as well as lecturer reflections on each cycle. Sample: Small-scale convenience samples of second- and third-year students studying in a Bachelor of Business degree, majoring in marketing, taking the Digital Promotions subject at a metropolitan university in Queensland, Australia participated in the study. The samples included students who had and had not experienced the field trip. The numbers of students taking part in the field trip ranged from 22 to 48 across the four semesters. Findings and Implications: The findings from the four iterations of the action research plan helped identify key considerations for incorporating technologies into learning environments. Feedback and reflections from the students and lecturer suggested that an innovative learning opportunity had been developed. However, pedagogical potential was limited, in part, by technological difficulties and by student perceptions of relevance.

Feature-domain super-resolution framework for gabor-based face and iris recognition

The low resolution of images has been one of the major limitations in recognising humans from a distance using their biometric traits, such as face and iris. Superresolution has been employed to improve the resolution and the recognition performance simultaneously, however the majority of techniques employed operate in the pixel domain, such that the biometric feature vectors are extracted from a super-resolved input image. Feature-domain superresolution has been proposed for face and iris, and is shown to further improve recognition performance by capitalising on direct super-resolving the features which are used for recognition. However, current feature-domain superresolution approaches are limited to simple linear features such as Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA), which are not the most discriminant features for biometrics. Gabor-based features have been shown to be one of the most discriminant features for biometrics including face and iris. This paper proposes a framework to conduct super-resolution in the non-linear Gabor feature domain to further improve the recognition performance of biometric systems. Experiments have confirmed the validity of the proposed approach, demonstrating superior performance to existing linear approaches for both face and iris biometrics.

Energy use, indoor temperature and possible adaptation strategies for air-conditioned office buildings in face of global warming

This paper discusses and summarises a recent systematic study on the implication of global warming on air conditioned office buildings in Australia. Four areas are covered, including analysis of historical weather data, generation of future weather data for the impact study of global warming, projection of building performance under various global warming scenarios, and evaluation of various adaptation strategies under 2070 high global warming conditions. Overall, it is found that depending on the assumed future climate scenarios and the location considered, the increase of total building energy use for the sample Australian office building may range from 0.4 to 15.1%. When the increase of annual average outdoor temperature exceeds 2 °C, the risk of overheating will increase significantly. However, the potential overheating problem could be completely eliminated if internal load density is significantly reduced.

Probabilistic matching of image sets for video-based face recognition

We address the problem of face recognition on video by employing the recently proposed probabilistic linear discrimi-nant analysis (PLDA). The PLDA has been shown to be robust against pose and expression in image-based face recognition. In this research, the method is extended and applied to video where image set to image set matching is performed. We investigate two approaches of computing similarities between image sets using the PLDA: the closest pair approach and the holistic sets approach. To better model face appearances in video, we also propose the heteroscedastic version of the PLDA which learns the within-class covariance of each individual separately. Our experi-ments on the VidTIMIT and Honda datasets show that the combination of the heteroscedastic PLDA and the closest pair approach achieves the best performance.

On the application of the probabilistic linear discriminant analysis to face recognition across expression

Facial expression is one of the main issues of face recognition in uncontrolled environments. In this paper, we apply the probabilistic linear discriminant analysis (PLDA) method to recognize faces across expressions. Several PLDA approaches are tested and cross-evaluated on the Cohn-Kanade and JAFFE databases. With less samples per gallery subject, high recognition rates comparable to previous works have been achieved indicating the robustness of the approaches. Among the approaches, the mixture of PLDAs has demonstrated better performances. The experimental results also indicate that facial regions around the cheeks, eyes, and eyebrows are more discriminative than regions around the mouth, jaw, chin, and nose.