Face recognition robust to head pose from one sample image

Most face recognition systems only work well under quite constrained environments. In particular, the illumination conditions, facial expressions and head pose must be tightly controlled for good recognition performance. In 2004, we proposed a new face recognition algorithm, Adaptive Principal Component Analysis (APCA) [4], which performs well against both lighting variation and expression change. But like other eigenface-derived face recognition algorithms, APCA only performs well with frontal face images. The work presented in this paper is an extension of our previous work to also accommodate variations in head pose. Following the approach of Cootes et al, we develop a face model and a rotation model which can be used to interpret facial features and synthesize realistic frontal face images when given a single novel face image. We use a Viola-Jones based face detector to detect the face in real-time and thus solve the initialization problem for our Active Appearance Model search. Experiments show that our approach can achieve good recognition rates on face images across a wide range of head poses. Indeed recognition rates are improved by up to a factor of 5 compared to standard PCA.

Effects of cation substitutions on the physical properties of M2M1T2O6 pyroxenes

In this PhD study, the effects of the cation substitutions on the physical properties of pyroxenes have been discussed. The results of this work extend the knowledge on pyroxenes with different chemical compositions. These properties might be used in the development of ceramic pigments, advanced materials and for the mineralogical phase identification. First of all, the crystallographic differences between Ge and Si pyroxenes have been examined. The structure of C2/c Ca rich Ge clinopyroxenes is very close to the low pressure C2/c structural configuration found in Ca-rich Si-pyroxenes. The shear of the unit cell is very similar, and the difference between a Ge end member and the corresponding Si-rich one is less than 1°. Instead, a remarkable difference exists between Ca-poor Si and Ge clinopyroxenes. First, Ca-poor Ge pyroxenes do not display a P21/c symmetry, but retain the C2/c symmetry; second, the observed C2/c structure shows, at room pressure, the configuration with highly kinked tetrahedral chains characteristic of the high pressure C2/c symmetry of Si Ca-poor pyroxenes. In orthopyroxenes, with Pbca symmetry, Ge-pyroxenes have volume larger than Si-pyroxenes. Samples along the system CaCoGe2O6 - CoCoGe2O6 have been synthesized at three different temperatures: 1050 °C, 1200 °C and 1250 °C. The aim of these solid state syntheses was to obtain a solid solution at ambient pressure, since the analogues Si-system needs high pressure. Unfortunately, very limited solution occurs because the structure forms of the two end member (high temperature for CaCoGe2O6 and high pressure CoCoGe2O6) are incompatible. The phase diagram of this system has been sketched and compared to that of Si. The cobalt end member (CoCoGe2O6) is stable at ambient pressure in two symmetries: at 1050 °C C2/c and 1200 °C Pbca. The impurity phase formed during these experiments is cobalt spinel. Raman spectroscopy has been used to investigate the vibrational properties of Ca-pyroxenes CaCoGe2O6, CaMgGe2O6, CaMgSi2O6 and CaCoSi2O6. A comparison between silicate and germanate pyroxenes shows significant changes in peak positions of the corresponding modes caused mainly by the difference of the Ge-Si atomic weight along with the distortion and compression of the coordination polyhedra. Red shift in Raman spectra of germanates has been calculated by a rough scale factor calculated by a simple harmonic oscillator model, considering the different bond lengths for 4-coordinated Si ~ 1.60- 1.65 Å vs Ge–O distance ~1.70 - 1.80 Å. The Raman spectra of CaMgGe2O6 and CaCoGe2O6 have been classified, in analogy with silicate (Wang et al., 2001) counterparts, in different ranges: - R1 (880-640 cm-1): strong T-O stretching modes of Ge and non-bridging O1 and O2 atoms within the GeO4 tetrahedron; - R2 (640-480 cm-1): stretching/bending modes of Ge-Obr-Ge bonds (chain stretching and chain bending); - R4 (480-360 cm-1): O-Ge-O vibrations; - R3 (360-240 cm-1): motions of the cations in M2 and M1 sites correlated with tetrahedral chain motion and tilting tetrahedra; - R5 (below 240 cm-1): lattice modes. The largest shift with respect to CaMgSi2O6 - CaCoSi2O6 is shown by the T-O stretching and chain modes. High-pressure Raman spectroscopy (up to about 8 GPa) on the same samples of Ca-pyroxenes using an ETH-type diamond anvil cell shows no phase transition within the P-ranges investigated, as all the peak positions vary linearly as a function of pressure. Our data confirm previous experimental findings on Si-diopside (Chopelas and Serghiou, 2000). In the investigated samples, all the Raman peaks shift upon compression, but the major changes in wavenumber with pressure are attributed to the chain bending (Ge-Obr-Ge bonds) and tetrahedra stretching modes (Ge-Onbr). Upon compression, the kinking angle, the bond lengths and T-T distances between tetrahedra decrease and consequently the wavenumber of the bending chain mode and tetrahedra stretching mode increases. Ge-pyroxenes show the higher P-induced peak-position shifts, being more compressible than corresponding silicates. The vibrational properties of CaM2+Ge2O6 (M2+ =Mg, Mn, Fe, Co, Ni, Zn) are reported for the first time. The wavenumber of Ge-Obr-Ge bending modes decreases linearly with increasing ionic radius of the M1 cation. No simple correlation has been found with M1 atomic mass or size or crystallographic parameters for the peak at ~850 cm-1 and in the low wavenumber regions. The magnetic properties of the system CaCoSi2O6 - CoCoSi2O6 have been investigated by magnetometry. The join is always characterized by 1 a.p.f.u. of cobalt in M1 site and this causes a pure collinear antiferromagnetic behaviour of the intra-chain superexchange interaction involving Co ions detected in all the measurements, while the magnetic order developed by the cobalt ions in M2 site (intra-chain) is affected by weak ferromagnetism, due to the non-collinearity of their antiferromagnetic interaction. In magnetically ordered systems, this non-collinearity effect promotes a spin canting of anti-parallel aligned magnetic moments and thus is a source of weak ferromagnetic behaviour in an antiferromagnetic. The weak ferromagnetism can be observed only for the samples with Co content higher than 0.5 a.p.f.u. in M2, when the concentration is sufficiently high to create a long range order along the M2 chain which is magnetically independent of M1 chain. The ferromagnetism was detected both in the M(T) at 10 Oe and M(H).

Effects of the build-up and resetting of auditory stream segregation on temporal discrimination

The tendency to hear a tone sequence as 2 or more streams (segregated) builds up, but a sudden change in properties can reset the percept to 1 stream (integrated). This effect has not hitherto been explored using an objective measure of streaming. Stimuli comprised a 2.0-s fixed-frequency inducer followed by a 0.6-s test sequence of alternating pure tones (3 low [L]-high [H] cycles). Listeners compared intervals for which the test sequence was either isochronous or the H tones were slightly delayed. Resetting of segregation should make identifying the anisochronous interval easier. The HL frequency separation was varied (0-12 semitones), and properties of the inducer and test sequence were set to the same or different values. Inducer properties manipulated were frequency, number of onsets (several short bursts vs. one continuous tone), tone:silence ratio (short vs. extended bursts), level, and lateralization. All differences between the inducer and the L tones reduced temporal discrimination thresholds toward those for the no-inducer case, including properties shown previously not to affect segregation greatly. Overall, it is concluded that abrupt changes in a sequence cause resetting and improve subsequent temporal discrimination. (PsycINFO Database Record © 2009 APA, all rights reserved)

Build-up of the tendency to segregate auditory streams:resetting effects of single deviant tones

The tendency to hear a sequence of alternating low (L) and high (H) frequency tones as two streams can be increased by a preceding induction sequence, even one composed only of same-frequency tones. Four experiments used such an induction sequence (10 identical L tones) to promote segregation in a shorter test sequence comprising L and H tones. Previous studies have shown that the build-up of stream segregation is usually reduced greatly when a sudden change in acoustic properties distinguishes all of the induction tones from their test-sequence counterparts. Experiment 1 showed that a single deviant tone, created by altering the final inducer (in frequency, level, duration, or replacement with silence) reduced reported segregation, often substantially. Experiment 2 partially replicated this finding, using changes in temporal discrimination as a measure of streaming. Experiments 3 and 4 varied the size of a frequency change applied to the deviant tone; the extent of resetting varied with size only gradually. The results suggest that resetting begins to occur once the change is large enough to be noticeable. Since the prior inducers always remained unaltered in the deviant-tone conditions, it is proposed that a single change actively resets the build-up evoked by the induction sequence.

A comparative study of 'joined-up' working in three regeneration programme case studies

This research concerns the development of coordination and co-governance within three different regeneration programmes within one Midlands city over the period from 1999 to 2002. The New Labour government, in office since 1997, had an agenda for ‘joining-up’ government, part of which has had considerable impact in the area of regeneration policy. Joining-up government encompasses a set of related activities which can include the coordination of policy-making and service delivery. In regeneration, it also includes a commitment to operate through co-governance. Central government and local and regional organisations have sought to put this idea into practice by using what may be referred to as network management processes. Many characteristics of new policies are designed to address the management of networks. Network management is not new in this area, it has developed at least since the early 1990s with the City Challenge and Single Regeneration Budget (SRB) programmes as a way of encouraging more inclusive and effective regeneration interventions. Network management theory suggests that better management can improve decision-making outcomes in complex networks. The theories and concepts are utilised in three case studies as a way of understanding how and why regeneration attempts demonstrate real advances in inter-organisational working at certain times whilst faltering at others. Current cases are compared to the historical case of the original SRB programme as a method of assessing change. The findings suggest that: The use of network management can be identified at all levels of governance. As previous literature has highlighted, central government is the most important actor regarding network structuring. However, it can be argued that network structuring and game management are both practised by central and local actors; Furthermore, all three of the theoretical perspectives within network management (Instrumental, Institutional and Interactive), have been identified within UK regeneration networks. All may have a role to play with no single perspective likely to succeed on its own. Therefore, all could make an important contribution to the understanding of how groups can be brought together to work jointly; The findings support Klijn’s (1997) assertion that the institutional perspective is dominant for understanding network management processes; Instrumentalism continues on all sides, as the acquisition of resources remains the major driver for partnership activity; The level of interaction appears to be low despite the intentions for interactive decision-making; Overall, network management remains partial. Little attention is paid to the issues of accountability or to the institutional structures which can prevent networks from implementing the policies designed by central government, and/or the regional tier.

Persistence in a random bond ising model of socio-econo dynamics

We study the persistence phenomenon in a socio-econo dynamics model using computer simulations at a nite temperature on hypercubic lattices in dimensions up to ve. The model includes a \social" local eld which contains the magnetization at time t. The nearest neighbour quenched interactions are drawn from a binary distribution which is a function of the bond concentration, p. The decay of the persistence probability in the model depends on both the spatial dimension and p. We nd no evidence of \blocking" in this model. We also discuss the implications of our results for possible applications in the social and economic elds. It is suggested that the absence, or otherwise, of blocking could be used as a criterion to decide on the validity of a given model in dierent scenarios.

May the contact lens force be with you

FULL TEXT: Like many people one of my favourite pastimes over the holiday season is to watch the great movies that are offered on the television channels and new releases in the movie theatres or catching up on those DVDs that you have been wanting to watch all year. Recently we had the new ‘Star Wars’ movie, ‘The Force Awakens’, which is reckoned to become the highest grossing movie of all time, and the latest offering from James Bond, ‘Spectre’ (which included, for the car aficionados amongst you, the gorgeous new Aston Martin DB10). It is always amusing to see how vision correction or eye injury is dealt with by movie makers. Spy movies and science fiction movies have a freehand to design aliens with multiples eyes on stalks or retina scanning door locks or goggles that can see through walls. Eye surgery is usually shown in some kind of day case simplified laser treatment that gives instant results, apart from the great scene in the original ‘Terminator’ movie where Arnold Schwarzenegger's android character encounters an injury to one eye and then proceeds to remove the humanoid covering to this mechanical eye over a bathroom sink. I suppose it is much more difficult to try and include contact lenses in such movies. Although you may recall the film ‘Charlie's Angels’, which did have a scene where one of the Angels wore a contact lens that had a retinal image imprinted on it so she could by-pass a retinal scan door lock and an Eddy Murphy spy movie ‘I-Spy’, where he wore contact lenses that had electronic gadgetry that allowed whatever he was looking at to be beamed back to someone else, a kind of remote video camera device. Maybe we aren’t quite there in terms of devices available but these things are probably not the behest of science fiction anymore as the technology does exist to put these things together. The technology to incorporate electronics into contact lenses is being developed and I am sure we will be reporting on it in the near future. In the meantime we can continue to enjoy the unrealistic scenes of eye swapping as in the film ‘Minority Report’ (with Tom Cruise). Much more closely to home, than in a galaxy far far away, in this issue you can find articles on topics much nearer to the closer future. More and more optometrists in the UK are becoming registered for therapeutic work as independent prescribers and the number is likely to rise in the near future. These practitioners will be interested in the review paper by Michael Doughty, who is a member of the CLAE editorial panel (soon to be renamed the Jedi Council!), on prescribing drugs as part of the management of chronic meibomian gland dysfunction. Contact lenses play an active role in myopia control and orthokeratology has been used not only to help provide refractive correction but also in the retardation of myopia. In this issue there are three articles related to this topic. Firstly, an excellent paper looking at the link between higher spherical equivalent refractive errors and the association with slower axial elongation. Secondly, a paper that discusses the effectiveness and safety of overnight orthokeratology with high-permeability lens material. Finally, a paper that looks at the stabilisation of early adult-onset myopia. Whilst we are always eager for new and exciting developments in contact lenses and related instrumentation in this issue of CLAE there is a demonstration of a novel and practical use of a smartphone to assisted anterior segment imaging and suggestions of this may be used in telemedicine. It is not hard to imagine someone taking an image remotely and transmitting that back to a central diagnostic centre with the relevant expertise housed in one place where the information can be interpreted and instruction given back to the remote site. Back to ‘Star Wars’ and you will recall in the film ‘The Phantom Menace’ when Qui-Gon Jinn first meets Anakin Skywalker on Tatooine he takes a sample of his blood and sends a scan of it back to Obi-Wan Kenobi to send for analysis and they find that the boy has the highest midichlorian count ever seen. On behalf of the CLAE Editorial board (or Jedi Council) and the BCLA Council (the Senate of the Republic) we wish for you a great 2016 and ‘may the contact lens force be with you’. Or let me put that another way ‘the CLAE Editorial Board and BCLA Council, on behalf of, a great 2016, we wish for you!’

Pain response and follow-up of patients undergoing panretinal laser photocoagulation with reduced exposure times

Purpose - We performed a study of laser panretinal photocoagulation in 20 patients with proliferative retinopathy. We compared short exposure, high-energy laser settings with conventional settings, using a 532?nm, frequency doubled, Neodymium–Yag laser and assessed the patients in terms of pain experienced and effectiveness of treatment. Methods - Twenty patients having panretinal photocoagulation for the first time underwent random allocation to treatment of the superior and inferior hemi-retina. Treatment A used ‘conventional’ parameters: exposure time 0.1?s, power sufficient to produce a visible grey-white burns, spot size 300?µm. The other hemi- retina was treated with treatment B using exposure 0.02?s, 300?µm and sufficient power to have similar endpoint. All patients were asked to evaluate severity of pain on a visual analogue scale. (0=no pain, 10=most severe pain). All patients were masked as to the type of treatment and the order of carrying out the treatment on each patient was randomised. Patients underwent fundus photography and were followed up for 6–45 months. Results - Seventeen patients had proliferative diabetic retinopathy, two had ischaemic central retinal vein occlusion and one had ocular ischaemic syndrome. The mean response to treatment A was 5.11, compared to 1.40 treatment B, on the visual analogue scale, which was statistically significant (P=0.001). All patients preferred treatment B. Further treatments, if required, were performed with treatment B parameters and long-term follow-up has shown no evidence of undertreatment. Conclusions - Shortening exposure time of retinal laser is significantly less painful but equally effective as conventional parameters.

Neonatal seizures and post-neonatal epilepsy:a seven-year follow-up study

BACKGROUND: Seizures are one of the most common symptoms of acute neurological disorders in newborns. This study aims at evaluating predictors of epilepsy in newborns with neonatal seizures. METHODS: we recruited consecutively eighty-five neonates with repeated neonatal video-EEG-confirmed seizures between Jan 1999 and Dec 2004. The relationship between clinical, EEG and ultrasound data in neonatal period and the development of post-neonatal epilepsy was investigated at 7 years of age. RESULTS: Fifteen patients (17.6%) developed post-neonatal epilepsy. Partial or no response to anticonvulsant therapy (OR 16.7, 95% CI: 1.8-155.8, p= .01; OR 47, 95% CI: 5.2-418.1, p<.01 respectively), severely abnormal cerebral ultrasound scan findings (OR: 5.4; 95% CI: 1.1-27.4; p<.04), severely abnormal EEG background activity (OR: 9.5; 95% CI: 1.6-54.2; p= .01) and the presence of status epilepticus (OR: 6.1; 95% CI: 1.8-20.3; p<.01) were found to be predictors of epilepsy. However, only the response to therapy seemed to be an independent predictor of post-neonatal epilepsy. CONCLUSION: Neonatal seizures seem to be related to post-neonatal epilepsy. Recurrent and prolonged neonatal seizures may act on an epileptogenic substrate, causing further damage, which is responsible for the subsequent clinical expression of epilepsy.

Build-up of auditory stream segregation induced by tone sequences of constant or alternating frequency and the resetting effects of single deviants

A sequence of constant-frequency tones can promote streaming in a subsequent sequence of alternating-frequency tones, but why this effect occurs is not fully understood and its time course has not been investigated. Experiment 1 used a 2.0-s-long constant-frequency inducer (10 repetitions of a low-frequency pure tone) to promote segregation in a subsequent, 1.2-s test sequence of alternating low- and high-frequency tones. Replacing the final inducer tone with silence substantially reduced reported test-sequence segregation. This reduction did not occur when either the 4th or 7th inducer was replaced with silence. This suggests that a change at the induction/test-sequence boundary actively resets build-up, rather than less segregation occurring simply because fewer inducer tones were presented. Furthermore, Experiment 2 found that a constant-frequency inducer produced its maximum segregation-promoting effect after only three tones—this contrasts with the more gradual build-up typically observed for alternating-frequency sequences. Experiment 3 required listeners to judge continuously the grouping of 20-s test sequences. Constant-frequency inducers were considerably more effective at promoting segregation than alternating ones; this difference persisted for ~10 s. In addition, resetting arising from a single deviant (longer tone) was associated only with constant-frequency inducers. Overall, the results suggest that constant-frequency inducers promote segregation by capturing one subset of test-sequence tones into an ongoing, preestablished stream, and that a deviant tone may reduce segregation by disrupting this capture. These findings offer new insight into the dynamics of stream segregation, and have implications for the neural basis of streaming and the role of attention in stream formation. (PsycINFO Database Record (c) 2013 APA, all rights reserved)

2-(3-Hydroxy-propyl)isoindoline-1,3-dione:Competition among hydrogen-bond acceptors

The title compound, C11H11NO3, has two mol-ecules in the asymmetric unit, which differ in the orientation of their side-chain OH groups, allowing them to form inter-molecular O - H⋯O hydrogen bonds to different acceptors. In one case, the acceptor is the OH group of the other mol-ecule, and in the other case it is an imide O=C group. This is the first example in the N-substituted phthalimide series in which independent mol-ecules have different types of acceptor. Mol-ecular-orbital calculations place the greatest negative charge on the OH group. © 2008 International Union of Crystallography.

Creating resilient SMEs:why one size might not fit all

This article is a first step towards addressing a gap in the field of organisational resilience research by examining how small and medium enterprises (SME) manage the threat and actuality of extreme events. Pilot research found that the managerial framing of extreme events varied by a range of organisational factors. This finding informed further examination of the contextual nature of the resilience concept. To date, large organisations have been the traditional focus of empirical work and theorising in this area; yet the heterogeneous SME sector makes up approximately 99% of UK industry and routinely operates under conditions of uncertainty. In a comparative study examining UK organisational resilience, it emerged that SME participants had both a distinctive perspective and approach to resilience when compared to participants from larger organisations. This article presents a subset of data from 11 SME decision-makers. The relationship between resilience capabilities, such as flexibility and adaptation, is interrogated in relation to organisational size. The data suggest limitations of applying a one-size-fits-all organisation solution (managerial or policy) to creating resilience. This study forms the basis for survey work examining the extent to which resilience is an organisationally contingent concept in practice.

Scaling up question-answering to linked data

Linked Data semantic sources, in particular DBpedia, can be used to answer many user queries. PowerAqua is an open multi-ontology Question Answering (QA) system for the Semantic Web (SW). However, the emergence of Linked Data, characterized by its openness, heterogeneity and scale, introduces a new dimension to the Semantic Web scenario, in which exploiting the relevant information to extract answers for Natural Language (NL) user queries is a major challenge. In this paper we discuss the issues and lessons learned from our experience of integrating PowerAqua as a front-end for DBpedia and a subset of Linked Data sources. As such, we go one step beyond the state of the art on end-users interfaces for Linked Data by introducing mapping and fusion techniques needed to translate a user query by means of multiple sources. Our first informal experiments probe whether, in fact, it is feasible to obtain answers to user queries by composing information across semantic sources and Linked Data, even in its current form, where the strength of Linked Data is more a by-product of its size than its quality. We believe our experiences can be extrapolated to a variety of end-user applications that wish to scale, open up, exploit and re-use what possibly is the greatest wealth of data about everything in the history of Artificial Intelligence. © 2010 Springer-Verlag.

The morphology of the labour party's one nation narrative:story, plot and authorship

By applying narrative theory to the party political texts emerging within the UK Labour Party after 2010, which make up the corpus of One Nation discourse, we can grasp the underlying significance of this ideational revision of Labour Party and leftist thought. Through an identification and analysis of the sequence of texts and their constitution as a "story" that interpolates an underlying "plot," we can see how a revision of Labour's "tale" offers to leadership a new party discourse appropriate to it, mediating-if not reconciling-the problematic duality of narrative authorship by both party and leader. © The Political Quarterly Publishing Co. Ltd. 2013.

Build-up of auditory stream segregation induced by tone sequences of constant or alternating frequency and the resetting effects of single deviants

A sequence of constant-frequency tones can promote streaming in a subsequent sequence of alternating-frequency tones, but why this effect occurs is not fully understood and its time course has not been investigated. Experiment 1 used a 2.0-s-long constant-frequency inducer (10 repetitions of a low-frequency pure tone) to promote segregation in a subsequent, 1.2-s test sequence of alternating low- and high-frequency tones. Replacing the final inducer tone with silence substantially reduced reported test-sequence segregation. This reduction did not occur when either the 4th or 7th inducer was replaced with silence. This suggests that a change at the induction/test-sequence boundary actively resets build-up, rather than less segregation occurring simply because fewer inducer tones were presented. Furthermore, Experiment 2 found that a constant-frequency inducer produced its maximum segregation-promoting effect after only three tones—this contrasts with the more gradual build-up typically observed for alternating-frequency sequences. Experiment 3 required listeners to judge continuously the grouping of 20-s test sequences. Constant-frequency inducers were considerably more effective at promoting segregation than alternating ones; this difference persisted for ~10 s. In addition, resetting arising from a single deviant (longer tone) was associated only with constant-frequency inducers. Overall, the results suggest that constant-frequency inducers promote segregation by capturing one subset of test-sequence tones into an ongoing, preestablished stream, and that a deviant tone may reduce segregation by disrupting this capture. These findings offer new insight into the dynamics of stream segregation, and have implications for the neural basis of streaming and the role of attention in stream formation. (PsycINFO Database Record (c) 2013 APA, all rights reserved)