Recognition from appearance subspaces across image sets of variable scale

Linear subspace representations of appearance variation are pervasive in computer vision. In this paper we address the problem of robustly matching them (computing the similarity between them) when they correspond to sets of images of different (possibly greatly so) scales. We show that the naïve solution of projecting the low-scale subspace into the high-scale image space is inadequate, especially at large scale discrepancies. A successful approach is proposed instead. It consists of (i) an interpolated projection of the low-scale subspace into the high-scale space, which is followed by (ii) a rotation of this initial estimate within the bounds of the imposed “downsampling constraint”. The optimal rotation is found in the closed-form which best aligns the high-scale reconstruction of the low-scale subspace with the reference it is compared to. The proposed method is evaluated on the problem of matching sets of face appearances under varying illumination. In comparison to the naïve matching, our algorithm is shown to greatly increase the separation of between-class and within-class similarities, as well as produce far more meaningful modes of common appearance on which the match score is based.

Freehand 3D Scanning in a Mobile Environment using Video

This work is motivated by two important trends in consumer computing: (i) the growing pervasiveness of mobile computing devices, and (ii) the users’ desire for increasingly complex but readily acquired and manipulated information content. Specifically, we develop and describe a system for 3D model creation of an object, using only a standard mobile device such as a smart phone. Our approach applies the structured light projection methodology and exploits multiple image input such as frames from a video sequence. In comparison with previous work, a significant further challenge addressed here is that of lower quality input data and limited hardware (processing power and memory, camera and projector quality). Novelties include: (i) a comparison of projection pattern detection approaches in the context of a mobile environment – a robust method combining colour detection and a phase congruency descriptor is evaluated, (ii) a model for single view reconstruction which exploits epipolar, coplanarity and topological constraints, (iii) the use of mobile device sensor data in the iterative closest point algorithm used to register multiple partial 3D reconstructions, and (iv) two heuristics for determining the order in which buffered single view based reconstructions are merged. Our experiments demonstrate that visually appealing results are obtained in a speedy manner which does not require specialist knowledge or expertise from the user.

Mixing Matrix Estimation From Sparse Mixtures With Unknown Number of Sources

In blind source separation, many methods have been proposed to estimate the mixing matrix by exploiting sparsity. However, they often need to know the source number a priori, which is very inconvenient in practice. In this paper, a new method, namely nonlinear projection and column masking (NPCM), is proposed to estimate the mixing matrix. A major advantage of NPCM is that it does not need any knowledge of the source number. In NPCM, the objective function is based on a nonlinear projection and its maxima just correspond to the columns of the mixing matrix. Thus a column can be estimated first by locating a maximum and then deflated by a masking operation. This procedure is repeated until the evaluation of the objective function decreases to zero dramatically. Thus the mixing matrix and the number of sources are estimated simultaneously. Because the masking procedure may result in some small and useless local maxima, particle swarm optimization (PSO) is introduced to optimize the objective function. Feasibility and efficiency of PSO are also discussed. Comparative experimental results show the efficiency of NPCM, especially in the cases where the number of sources is unknown and the sources are relatively less sparse.

Chime

A multi-disciplinary, site-specific performance which moved between several spaces in the Courthouse Arts building. The work combined improvised dance, choreographic practice, video projection and sound which utilized and computer-manipulated the Federation Handbells.

Presents//presence: Telematics, performance and embodiment

Of giving and receiving // Of giving and taking // Of exchanges
Reciprocity // Mutuality // Expectations

A performance of liveness in which the presence of the performer is interrogated.

Drawing on Dan Graham's (1974) Two Consciousness Project, Presents//Presence plays on Object = Space relationships. Engaging with contemporary notions of thinking and consciousness, the performance plays with time - the here and now, the passage of time, time zones, and timing.

Lovers. An anniversary. Fine Dining. Distance. Skype.

Presents disrupt subject positions of audience, of performer; something happens while we are waiting for something else to happen. Through the use portable computers and hand-held (smart) devices for the capture and 'projection' of action in real time, the exploration sets out to engage with notions live and remote, absence and presence, the play of embodied transmission and live performance and the perception of absence.

Through a simulation of the simultaneous presences (performances) of performers Magda Miranda and Rea Dennis, Presents//Presence is a performative event that (re) activates live/d moments in the lives of the artists. Such presence characterizes computer time – “a permanent present, an unbounded, timeless intensity” (Virilio in Dixon 90). Such presence could also be said to characterize the intra-subjective experience of intimacy. The piece draws on the languages of live theatre, elements of autobiographical performance, inter- and intra-subjective perception, and an understanding of time as a spatial metaphor. This paper reports on the performance event Presents//Presence. In it, we outline the narrative and structural anchors that frame the piece and discuss some of the theoretical threads informing the research. The paper is accompanied by a recording of the performance that was delivered to a live audience at University of South Wales, Cardiff during the Remote Encounters conference in April 2013.

When i look outside the window my shoes creak

 A solo exhibition consisting of a video projection

Conversation

 'Conversation' was part of Melbourne Urban Screens Festival at Federation Square from 3 - 5 Oct. 2008 and Rotterdam International Film Festival (38th : 2009 : Rotterdam, The Netherlands) from 21 Jan. - 1 Feb. 2009. Both events curated by Mirjam Struppek

Forest of gestures : video installation and performance work

Invitation to contribute to Alexander Harrisions' program 'What's Coming' resulted in 'Forest of Gestures' a 3 channel video projection and installation with a live feed component

Frayed

A solo exhibition using large scale video projection and sound with accompanying catalogue 'Semblance at the Top' by Melissa Amore

Learning user preference patterns for Top-N recommendations

In this paper, we observe that the user preference styles tend to change regularly following certain patterns. Therefore, we propose a Preference Pattern model to capture the user preference styles and their temporal dynamics, and apply this model to improve the accuracy of the Top-N recommendation. Precisely, a preference pattern is defined as a set of user preference styles sorted in a time order. The basic idea is to model user preference styles and their temporal dynamics by constructing a representative subspace with an Expectation- Maximization (EM)-like algorithm, which works in an iterative fashion by refining the global and the personal preference styles simultaneously. Then, the degree which the recommendations match the active user's preference styles, can be estimated by measuring its reconstruction error from its projection on the representative subspace. The experiment results indicate that the proposed model is robust to the data sparsity problem, and can significantly outperform the state-of-the-art algorithms on the Top-N recommendation in terms of accuracy. © 2012 IEEE.

Influence functions of two families of robust estimators under proportional scatter matrices

In this paper, under a proportional model, two families of robust estimates for the proportionality constants, the common principal axes and their size are discussed. The first approach is obtained by plugging robust scatter matrices on the maximum likelihood equations for normal data. A projection- pursuit and a modified projection-pursuit approach, adapted to the proportional setting, are also considered. For all families of estimates, partial influence functions are obtained and asymptotic variances are derived from them. The performance of the estimates is compared through a Monte Carlo study. © 2006 Springer-Verlag.