Super-resolution of a 3-dimensional scene from novel viewpoints

Super-resolution is a method of post-processing image enhancement that increases the spatial resolution of video or images. Existing super-resolution techniques apply only to images captured of a planar scene. This paper aims to extend super-resolution concepts from the 2D domain to the 3D domain, drawing on ideas from both superresolution and multi-view geometry, two fields of research that until now have predominantly been studied in isolation. 2D super-resolution methods are not without their complexities and limitations. However, once multiple views of a scene are considered within a super-resolution framework, a new range of issues arise that must also be resolved. For example, when input images of a scene with variation in depth are considered, it is no longer clear how and where the images should be registered. This paper describes the use of sparse 3D reconstruction in order to ‘register’ the input images, which are then transferred to a novel image plane and combined to increase the perceived detail in the scene. Experimental results using real images captured from generally positioned input cameras are presented.

Gould's book of fish : a novel in 12 fish (Review)

Book review of "Gould’s book of fish : a novel in 12 fish" by Richard Flanagan

Identification and characterisation of novel genes involved in skeletal muscle hypertrophy

There is mounting evidence in support of the view that skeletal muscle hypertrophy results from the complex and coordinated interaction of numerous signalling pathways. Well characterised components integral to skeletal muscle adaptation include the transcriptional activity of the members of the myogenic regulatory factors, numerous secreted peptide growth factors, and the regenerative potential of satellite cells. Whilst studies investigating isolated components or pathways have enhanced our current understanding of skeletal muscle hypertrophy, our knowledge of how all of these components react in concert to a common stimulus remains limited. The broad aim of this thesis was to identify and characterise novel genes involved in skeletal muscle hypertrophy. We have created a customised human skeletal muscle specific microarray which contains ∼11,000 cDNA clones derived from a normalised human skeletal muscle cDNA library as well as 270 genes with known functional roles in human skeletal muscle. The first aspect of this thesis describes the production of the microarray and evaluates the robustness and reproducibility of this analytical technique. Study one aimed to use this microarray in the identification of genes that are differentially expressed during the forced differentiation of human rhabdomyosarcoma cells, an in vitro model of skeletal muscle development. Firstly using this unique model of aberrant myogenic differentiation we aimed to identify genes with previously unidentified roles in myogenesis. Secondly, the data from this study permitted the examination of the performance of the microarray in detecting differential gene expression in a biological system. We identified several new genes with potential roles in the myogenic arrest of rhabdomyosarcoma and further characterised the expression of muscle specific genes in rhabdomyosarcoma differentiation. In study two, the molecular responses of cell cycle regulators, muscle regulatory factors, and atrophy related genes were mapped in response to a single bout of resistance exercise in human skeletal muscle. We demonstrated an increased expression of MyoD, myogenin and p21, whilst the expression of myostatin was decreased. The results of this study contribute to the existing body of knowledge on the molecular regulation skeletal muscle to a hypertrophic stimulus. In study three, the muscle samples collected in study two were analysed using the human skeletal muscle specific microarray for the identification of novel genes with potential roles in the hypertrophic process. The analysis uncovered four interesting genes (TXNIP, MLP, ASB5, FLJ 38973) that have not previously been examined in human skeletal muscle in response to resistance exercise. The functions of these genes and their potential roles in skeletal muscle are discussed. In study four, the four genes identified in study three were examined in human primary skeletal muscle cell cultures during myogenic differentiation. Human primary skeletal muscle cells were derived from the vastus lateralis muscle of 8 healthy volunteers (6 males and 2 females). Cell cultures were differentiated using serum withdrawal and serum withdrawal combined with IGF-1 supplementation. Markers of the cell proliferation, cell cycle arrest and myogenic differentiation were examined to assess the effectiveness of the differentiation stimulus. Additionally, the expressions of TXNIP, MLP, ASB5 and FLJ 38973 measured in an attempt to characterise further their roles in skeletal muscle. The expression of TXNIP changed markedly in response to both differentiation stimuli, whilst the expression of the remaining genes were not altered. Therefore it was suggested that expression of these genes might be responsive to the mechanical strain or contraction induced by the resistance exercise. In order to examine whether these novel genes responded specifically to resistance type exercise, their expression was examined following a single bout of endurance exercise. The expression of TXNIP, MLP, and FLJ 38973 remained unchanged whilst ASB5 increased 30 min following the cessation of the exercise.

Second person : a point of view? The function of the second person pronoun in narrative prose fiction

This thesis looks at the functions and effects of the ‘second-person’ pronoun in narrative prose fiction, with particular focus on the fluidity and ambiguity of the mode that I will call Protean-'you.' It is a mode in which it is unclear whether the ‘you’ is a character, the narrator, a reader/narratee, or no-one in particular—or a combination of these—so that readers find ‘second-person’ utterances at once familiar and deeply strange. I regard the ‘second person’ as a special case of narrative ‘person’ that, at its most fluid, can produce an experience of reading quite unlike that of reading traditional ‘first-‘ and ‘third-person’ narrative. Essentially, this unique experience comes about because Protean-‘you’ neglects to constitute the stable modes of subjectivity that readers expect to find within narrative textuality. These stable modes of subjectivity, modelled on what I will refer to as Cartesianism’s hegemonic notion of the self, have been thoroughly formalised and naturalised within the practices of ‘first-‘ and ‘third-person’ narrative. The Protean-‘you’ form of ‘second-person’ narrative, conversely, is a mode of narrative discourse that puts readers in a place of doubt and uncertainty, its unsettling equivocations forcefully disrupting accustomed, mimetic explanations of narrative and denying us access to the foundational, authorising subject of classical Cartesian thought. Rather than founding a notion of ‘second-person’ narrative and narrative ‘person’ generally on Cartesianism's ‘self-ish’ logic of unified, privatised identity, I turn to C.S, Peirce's notion of the semiotic self and to developments in post-structuralist thought. Essentially, the conception of subjectivity underpinning my arguments is Peirce's proposition that the self is to be conceived of not as a cogito, but as a sign by which the conscious entity knows itself. It is a sign, moreover, that is constantly being re-read, reinterpreted, so that identity is never self-complete. This reconception of subjectivity is necessary because 1 will argue that the effects of Protean-‘you’ arise in some part from a tension between Cartesianism's hegemony and what philosophical pragmatism and post-structuralism glimpse as the actual condition of the human subject—the subject as dispersed and contingent rather than unified and authoritative. Most discussions of ‘second-person’ narrative conceive of the mode in terms of implicit communicative relations, in some measure instituting Cartesianism's notion of the intentionalist self at the centre of literary meaning. I contrast the paradigmatic address model that arises from this conception against a model that approaches the analysis of ‘second-person’ narrative modality in terms of a referential function, that is, in terms of the object or objects referred to deictically by the ‘second-person’ pronoun. Two principal functions of ‘second-person’ textuality are identified and discussed at length. The first is generalisation, which is rarely dissipated altogether, a situation that contributes to the ambiguities of the pronoun's reference in much ‘second-person’ fiction. The second principal function is that of address, that is, the allocutionary function. Clearly, although stories that continually refer to a ‘you’ can seem quite baffling and unnatural, not all ‘second-person’ narratives unsettle the reader. In order to make the ‘second person's’ outlandish narratives knowable and stable, we bring to bear on them in our habits of reading whatever hermeneutic frames, whatever interpretive keys, come to hand, including a large number of unexceptional forms of literary and ‘natural’ discourse that employ the ‘second-person’ pronoun. These forms include letter writing and internal dialogue (i.e., talking to one's self), the language of the courtroom, the travelogue, the maxim, and so on. In looking at the ways in which the radicalising potentials of ‘second-person’ discourse are contained or recuperated, I focus on issues of vraisemblance and mimesis. Vraisemblance can be described as the ‘system of conventions and expectations which rests on/reinforces that more general system of ‘mutual knowledge’ produced within a community for the realisation and maintenance of a whole social world’. All of the forms of the vraisemblable are already instituted within social, cultural relations, so that what vraisemblance describes is the way we fit the inscriptions we read-that is, the way in which we naturalise what we read-into those given cultural and social forms. I also look at the conventionalising and naturalising work done by notions of mimesis in explaining relations between the world, our being in it, and texts, proposing that mimesis provides a principle buttress by which the good standing of the metaphor of ‘person’ is preserved in traditional and pre-critical modes of analysis. Indeed, the critic’s recourse to ‘person’ is in some measure always an engagement with mimesis. Any discussion that maintains that mimesis is in some way productive of meaning-which this thesis in fact does-must identify mimesis as a merely conventional category within practices of reading and semiosis more generally, and at the very least remove that term from its traditional position of transparent primacy and authority. Some of the most interesting and insightful arguments about ‘second-person’ narrative propose that the ‘second person’s’ most striking effects derive from the constitution of an ‘intersubjective’ experience of reading in which the subject positions of the ‘you’-protagonist, reader-narratee and narrator are combined into a fluid and indeterminate multiple subjectivity. Notions of intersubjectivity frequently position themselves as liberating the reader from Cartesianism's fixed, authoritative modes of subjectivity, Frequently, however, they tend implicitly to reinstate Cartesianism's notion of the self at the centre of textual practice and subjectivity. I look at Daniel Gunn's novel ‘Almost You’, at length in this context, illustrating the constant overdetermination of the ‘you’ and the novel's narrating voice, and demonstrating that this overdetermination leaves the origin of the narrative discourse, the identity of the narrator, and the ontological nature of both principal protagonists utterly ambiguous. The fluidity and ambiguity of Protean-‘you’ in ‘Almost You’ is discussed in terms of ‘second-person’ intersubjectivity, but with a view to demonstrating the indebtedness by the notion of intersubjectivity to Cartesianism's hegemony of ‘person’. I then turn to a discussion of what might be a more ‘old fashioned’ if perhaps ultimately more far-reaching approach to the ‘second person’s’ often startling ambiguities. This is Keats's notion of negative capability, a capacity or quality in which a person ‘is capable of being in uncertainties, Mysteries, doubts, without any irritable reaching after fact and reason.’ I suggest that Protean-‘you’ texts will license all of the readings of ambiguity and fluidity proposed in my discussion of ‘Almost You’, but conclude that the instances of indeterminacy illustrate no more than that: the fluidity and deep ambiguity, and thus, finally, the lack of coherence, of Protean-‘you’ discourse. This has particular implications for how we are to consider readers’ experiences of narrative texts. More fundamentally, it has implications for how we are to consider readers as subjects. I suggest that unstable, ambiguous instances of ‘second-person’ narrative can tear the complex and systematic embroidery of ideological suture that unifies Cretinism’s experience or sense of subjectivity, leaving the reader in a condition of epistemological and ontological havoc. I go on to argue that much of the deeply unsettling effect of Protean-‘you’ discourse anises because its utterances explicitly gesture towards Cretinism’s notion of self. Protean-‘you’ involves a sense of address that is much more pronounced than we are accustomed to facing when reading literary narrative, alerting us to the presence of inscribed anthropomorphic subjects. At the very same time, protean-‘you’ leaves its inscribed subjects indeterminate, ambiguous. This conflict generates a tension between the anticipation of the emergence of speaking and listening selves and our inability to find them. I go on to propose that Protean-‘you’ narrative's lack of coherence is also to be understood as the condition of narrative actuality generally, but a condition that is vigorously mediated against by dominant practices of reading and writing, hocusing my discussion in this respect on the issue of narrative ‘person,’ I argue that narrative ‘person’ is constituted within texts as an apparent unity, but that it is in fact, produced as unitary solely within the practice of making sense, that is, Within our habits of reading, and so is never finally unified. I propose that this is the case for ‘first-‘ and ‘third-person’ modes no less than for the ‘second.’ Where ‘second-person’ narrative at its most radical and Protean differs from conventional ‘first-‘ and ‘third-person’ narratives is the degree to which each has been circumscribed by practices of tantalization, containment and limit, and, in particular, Cretinism’s hegemony of ‘person.’ It may be that the most significant insights ‘second-person’ narrative has to offer are to be found within its capacity to reveal to the engaged reader the underlying condition of narrative discourse, and more generally, its capacity to reveal the actual condition of the human subject-a condition in which, exactly like its textual corollary of narrative ‘person,’ the self is glimpsed as thoroughly dispersed and contingent.

CAFS : a novel lightweight cache-based scheme for large-scale intrusion alert fusion

In this paper, we present some practical experiences on implementing an alert fusion mechanism from our project. After investigation on most of the existing alert fusion systems, we found the current body of work alternatively weighed down in the mire of insecure design or rarely deployed because of their complexity. As confirmed by our experimental analysis, unsuitable mechanisms could easily be submerged by an abundance of useless alerts. Even with the use of methods that achieve a high fusion rate and low false positives, attack is also possible. To find the solution, we carried out analysis on a series of alerts generated by well-known datasets as well as realistic alerts from the Australian Honey-Pot. One important finding is that one alert has more than an 85% chance of being fused in the following five alerts. Of particular importance is our design of a novel lightweight Cache-based Alert Fusion Scheme, called CAFS. CAFS has the capacity to not only reduce the quantity of useless alerts generated by intrusion detection system, but also enhance the accuracy of alerts, therefore greatly reducing the cost of fusion processing. We also present reasonable and practical specifications for the target-oriented fusion policy that provides a quality guarantee on alert fusion, and as a result seamlessly satisfies the process of successive correlation. Our experiments compared CAFS with traditional centralized fusion. The results showed that the CAFS easily attained the desired level of simple, counter-escapable alert fusion design. Furthermore, as a lightweight scheme, CAFS can easily be deployed and excel in a large amount of alert fusions, which go towards improving the usability of system resources. To the best of our knowledge, our work is a practical exploration in addressing problems from the academic point of view. Copyright © 2011 John Wiley & Sons, Ltd.

Mixed-norm sparse representation for multi view face recognition

Face recognition with multiple views is a challenging research problem. Most of the existing works have focused on extracting shared information among multiple views to improve recognition. However, when the pose variation is too large or missing, 'shared information' may not be properly extracted, leading to poor recognition results. In this paper, we propose a novel method for face recognition with multiple view images to overcome the large pose variation and missing pose issue. By introducing a novel mixed norm, the proposed method automatically selects candidates from the gallery to best represent a group of highly correlated face images in a query set to improve classification accuracy. This mixed norm combines the advantages of both sparse representation based classification (SRC) and joint sparse representation based classification (JSRC). A trade off between the ℓ1-norm from SRC and ℓ2,1-norm from JSRC is introduced to achieve this goal. Due to this property, the proposed method decreases the influence when a face image is unseen and has large pose variation in the recognition process. And when some face images with a certain degree of unseen pose variation appear, this mixed norm will find an optimal representation for these query images based on the shared information induced from multiple views. Moreover, we also address an open problem in robust sparse representation and classification which is using ℓ1-norm on the loss function to achieve a robust solution. To solve this formulation, we derive a simple, yet provably convergent algorithm based on the powerful alternative directions method of multipliers (ADMM) framework. We provide extensive comparisons which demonstrate that our method outperforms other state-of-the-arts algorithms on CMU-PIE, Yale B and Multi-PIE databases for multi-view face recognition.

Multi-view subspace clustering for face images

In many real-world computer vision applications, such as multi-camera surveillance, the objects of interest are captured by visual sensors concurrently, resulting in multi-view data. These views usually provide complementary information to each other. One recent and powerful computer vision method for clustering is sparse subspace clustering (SSC); however, it was not designed for multi-view data, which break down its linear separability assumption. To integrate complementary information between views, multi-view clustering algorithms are required to improve the clustering performance. In this paper, we propose a novel multi-view subspace clustering by searching for an unified latent structure as a global affinity matrix in subspace clustering. Due to the integration of affinity matrices for each view, this global affinity matrix can best represent the relationship between clusters. This could help us achieve better performance on face clustering. We derive a provably convergent algorithm based on the alternating direction method of multipliers (ADMM) framework, which is computationally efficient, to solve the formulation. We demonstrate that this formulation outperforms other alternatives based on state-of-The-Arts on challenging multi-view face datasets.