Virtual Visual Hulls: Example-Based 3D Shape Estimation from a Single Silhouette

Recovering a volumetric model of a person, car, or other object of interest from a single snapshot would be useful for many computer graphics applications. 3D model estimation in general is hard, and currently requires active sensors, multiple views, or integration over time. For a known object class, however, 3D shape can be successfully inferred from a single snapshot. We present a method for generating a ``virtual visual hull''-- an estimate of the 3D shape of an object from a known class, given a single silhouette observed from an unknown viewpoint. For a given class, a large database of multi-view silhouette examples from calibrated, though possibly varied, camera rigs are collected. To infer a novel single view input silhouette's virtual visual hull, we search for 3D shapes in the database which are most consistent with the observed contour. The input is matched to component single views of the multi-view training examples. A set of viewpoint-aligned virtual views are generated from the visual hulls corresponding to these examples. The 3D shape estimate for the input is then found by interpolating between the contours of these aligned views. When the underlying shape is ambiguous given a single view silhouette, we produce multiple visual hull hypotheses; if a sequence of input images is available, a dynamic programming approach is applied to find the maximum likelihood path through the feasible hypotheses over time. We show results of our algorithm on real and synthetic images of people.

Learning and Example Selection for Object and Pattern Detection

This thesis presents a learning based approach for detecting classes of objects and patterns with variable image appearance but highly predictable image boundaries. It consists of two parts. In part one, we introduce our object and pattern detection approach using a concrete human face detection example. The approach first builds a distribution-based model of the target pattern class in an appropriate feature space to describe the target's variable image appearance. It then learns from examples a similarity measure for matching new patterns against the distribution-based target model. The approach makes few assumptions about the target pattern class and should therefore be fairly general, as long as the target class has predictable image boundaries. Because our object and pattern detection approach is very much learning-based, how well a system eventually performs depends heavily on the quality of training examples it receives. The second part of this thesis looks at how one can select high quality examples for function approximation learning tasks. We propose an {em active learning} formulation for function approximation, and show for three specific approximation function classes, that the active example selection strategy learns its target with fewer data samples than random sampling. We then simplify the original active learning formulation, and show how it leads to a tractable example selection paradigm, suitable for use in many object and pattern detection problems.

First example of asymmetric transfer hydrogenation in water induced by a chiral amino alcohol hydrochloride

The low-cost and commercially available (-)-ephedrine hydrochloride was firstly employed in the [RuCl2(1)-cymene)](2)-catalyzed asymmetric transfer hydrogenation of prochiral ketones in water. The reaction could be performed in the open air at rt, affording excellent yields (up to 99%) and good enantioselectivities (up to 83% ee). It provided a further step toward the discovery of simplified catalyst systems for eventual availability. (c) 2005 Published by Elsevier Ltd.

Robotic Assembly of Naturally Varying Food Items via Teaching by Example

T. G. Williams, J.J. Rowland, and Lee M.H., Robotic Assembly of Naturally Varying Food Items via Teaching by Example, 9th Int. Symp. on Intelligent Robotic Systems (SIRS 2001), July 2001, France, pp133-142.

Teaching by Example in Food Assembly by Robot

T. G. Williams, J.J. Rowland, Lee M.H. and M.J. Neal Teaching by Example in Food Assembly by Robot, Proc. 2000 IEEE Int. Conf. On Robotics and Automation, San Francisco, April 2000, pp3247-52.

Rola Unii Europejskiej w budowie nowego ładu międzynarodowego - stan obecny i perspekytywy

W nauce o stosunkach międzynarodowych porządek, czy też ład międzynarodowy, to zbiór podmiotów państwowych i niepaństwowych złączonych wzajemnymi relacjami o różnorodnym charakterze i działających według wspólnie wypracowanych lub narzuconych z zewnątrz reguł. Po zakończeniu zimnej wojny i utracie na znaczeniu ładu jałtańskiego rozgorzała dyskusja nad kształtem jaki ma przybrać ład postzimnowojenny. Z poczynionych rozważań wynika, że polityczny i gospodarczy ład międzynarodowy w najbliższych dekadach będzie wielobiegunowy. Szczególne znaczenie odgrywać w nim będą USA, Chiny , a także Unia Europejska, Japonia, Indie i Brazylia. Obecnie UE odgrywa jedną z najważniejszych ról w budowie ładu międzynarodowego. Widać to zwłaszcza pod względem gospodarczym jako że Unia UE jest obecnie największym światowym eksporterem i drugim (po USA) importerem. Wydaje się jednak, że tradycyjne opisywanie Unii Europejskiej jako przykładu obszaru sukcesu gospodarczego może w najbliższym czasie ulec zmianie. Europa w długiej perspektywie ulegnie marginalizacji jeśli nie wróci do projektów ściślejszej integracji, nie wykorzysta lepiej swojego potencjału intelektualnego i gospodarczego oraz nie przyspieszy rozwoju poprzez stawianie na nowe technologie. Dla podtrzymania swojej pozycji powinna ona aktywniej współpracować z USA, utrzymywać poprawne stosunki z krajami azjatyckimi a także zaangażować się we współpracę z organizacjami między- i pozarządowymi, które w najbliższych latach będą zyskiwać na znaczeniu

Market allocations of location choice: an example

The purpose of this paper is to make an example which, first, illustrates Starret’s Spatial Imposibility Theorem, when agents have free mobility; and second, allowes us to get a competitive equilibrium with transportation when agents move only if there is a noticeable difference in utilities that justifies the change of location.

Example-Based Image Restoration via Boosted Classifiers

We propose a novel image registration framework which uses classifiers trained from examples of aligned images to achieve registration. Our approach is designed to register images of medical data where the physical condition of the patient has changed significantly and image intensities are drastically different. We use two boosted classifiers for each degree of freedom of image transformation. These two classifiers can both identify when two images are correctly aligned and provide an efficient means of moving towards correct registration for misaligned images. The classifiers capture local alignment information using multi-pixel comparisons and can therefore achieve correct alignments where approaches like correlation and mutual-information which rely on only pixel-to-pixel comparisons fail. We test our approach using images from CT scans acquired in a study of acute respiratory distress syndrome. We show significant increase in registration accuracy in comparison to an approach using mutual information.

Rallying to a cause: the plays of Zakes Mda 1979-1989

Zakes Mda, dubbed one of South Africa's most prolific playwrights, produced his richest and most powerful theatre work during the 70s and 80s. Ironically, it is only in the 90s that he has been acknowledged in his own country as one of its foremost dramatists - ironic since he has recently moved away from drama into the realms of fiction. Fortunately Mda has accumulated a worthy canon of dramatic works, spanning radio and film, as well as theatre, and there is no reason to believe that he will not return to play writing. Mda has worked extensively in theatre in various capacities but most notably in the area of theatre-for-development. For example, he worked as director with Maratholi Travelling Theatre in Lesotho, an experience which contributed, in part, towards his book "When People Play People: Development Communication Through Theatre". Mda's plays have been produced in the United States, Britain, Spain, France and Russia as well as in southern Africa. "The Nun's Romantic Story" has been translated into Castilian and Catalan and "We Shall Sing for the Fatherland" and "Dark Voices Ring" have both been translated into Russian and French. In South Africa he won the Merit Award of the Amstel Playwright of the Year Society for "We Shall Sing for the Fatherland" in 1978 and in 1979 he was Amstel Playright of the Year for "The Hill". For his novel "She Plays with the Darkness", he won the Sanlam Literary Award in 1995.

Invariant measure selection by noise. An example

We consider a deterministic system with two conserved quantities and infinity many invariant measures. However the systems possess a unique invariant measure when enough stochastic forcing and balancing dissipation are added. We then show that as the forcing and dissipation are removed a unique limit of the deterministic system is selected. The exact structure of the limiting measure depends on the specifics of the stochastic forcing.

WORKS BY FRANCO-BELGIAN COMPOSERS AND THE VIOLINISTS WHO INSPIRED THEM

In the late nineteenth century, French composers such as Camille Saint- Saens, Cesar Franck, and Claude Debussy worked to elevate instrumental music in late-Romantic period France, creating symphonies, concertos, and chamber ensembles, including duo sonatas. These composers and followers like, Ernest Chausson and Guillaume Lekeu were all influenced by a particular violinist to whom they dedicated their compositions. The primary violinist who inspired these composers was Eugene Ysaye (1858-1931), a brilliant performer and composer. His freedom of expression motivated many prominent French composers to dedicate major works to him. For example, Debussy dedicated his string quartet to Ysaye, who established the Ysaye Quartet and premiered Debussy's composition. In 1886, Franck completed his sonata for violin and piano which he also dedicated to Ysaye. Fritz Kreisler (1875-1962), one of the most talented violinists of his era, had a relationship withYsaye that was quite special. They respected, supported, and befriended each other. To Ysaye, Kreisler dedicated his Recitativo and Scherzo. To Kreisler, Ysaye dedicated one ofhis celebrated Sonatas for Solo Violin. Pablo de Sarasate (1844-1908) was a magnificent Spanish violinist of the late nineteenth century, and his music and performances influenced many composers, especially Saint-Saens, who included Spanish gypsy fragments in his works. These motifs may found in his Havanaise, Introduction and Rondo Capriccioso and Violin Concerto No.3 which were dedicated to Sarasate. My goal for this dissertation project has been to find and present, in three recitals, works by French composers and also works by the violinists who inspired them. As a violinist, I have endeavored to understand the influence of the various violinists on these French composers and how that knowledge can inform my approach to performing these works. In my first recital, with pianist Soo Young Jung, I performed works by Saint-Saens, Ysaye and Sarasate. With pianist Sun Ha Yoon, I performed works by Ysaye, Debussy, Kreisler and Franck in my second recital. My third recital, again with pianist Sun Ha Yoon, featured works by Ysaye, Chausson, and Lekeu. All recitals were recorded and performed at the University ofMaryland, College Park.

Risikowahrnehmung und psychologische Reaktionen in öffentlichen Krisensituationen am Beispiel von Terrorattentaten / Risk perception and psychological reactions in public crisis situations using the example of terror attacks

The reactions to the 9/11 terror attacks were immense in the western population. In the current review, the impact of terror attacks is presented with surveys, clinical interviews, and scientific polls, which were identified in a comprehensive literature search. Results show that the fear of further terror attacks is comparatively overestimated in the population and is associated with numerous psychological consequences and reactions. The overestimation of the probability of further terror attacks is related among other reasons to its unique features and its strong representation in the media. Several independent studies proved that the number of stress symptoms and psychiatric diagnoses is associated with a high risk perception in relation to terror attacks. This was not only the case for victims of terror attacks, but also for people indirectly exposed to the terror attacks. In addition, there is evidence that the number of the stress symptoms correlate with the duration of TV consumption of new findings about terror attempts. Methodologically, there is a critical lack of more in-depth analyses to explain the development of risk perceptions and its influence on mental and physical health. Because of the international importance and cross-cultural differences, an international standardization of research is desirable. [In German] Die Reaktionen auf die Terrorattentate vom 9. September 2001 in New York waren in der westlichen Bevölkerung immens. In der vorliegenden Übersichtsarbeit werden die Auswirkungen von Terrorattentaten durch Einbeziehung bevölkerungsrepräsentativer Untersuchungen, Surveys, klinischer Interviews und Einstellungsbefragungen dargestellt, die über eine deskriptive Literaturrecherche ermittelt wurden. Als Ergebnis des Reviews zeigt sich, dass die Angst vor weiteren Terrorattentaten in der Bevölkerung vergleichsweise hoch und mit zahlreichen psychologischen Folgen und Reaktionen assoziiert ist. Die Einschätzung der Auftretenswahrscheinlichkeit eines Terrorattentats hängt unter anderem mit den besonderen Charakteristika und der hohen medialen Präsenz des Themas zusammen. Die Anzahl der Stresssymptome bis hin zu psychiatrischen Diagnosen erwies sich in mehreren unabhängigen Untersuchungen mit einer hohen Risikowahrnehmung assoziiert. Dies ließ sich nicht nur bei den Opfern von Terrorattentaten, sondern auch bei indirekt Betroffenen zeigen. Darüber hinaus gibt es mehrfache Belege dafür, dass die Anzahl der Stresssymptome mit der Dauer des TV-Konsums über Neuigkeiten zu Terrorattentaten zusammenhing. Als methodische Kritik ist an den gegenwärtigen Untersuchungsszenarien einzuwenden, dass es derzeit keine tiefer gehenden Analysen zur Entwicklung der Risikowahrnehmung und zu ihrem Einfluss auf die Gesundheit gibt. Aufgrund der internationalen Bedeutung des Themas und der interkulturellen Unterschiede im Umgang mit Krisensituationen ist eine internationale Standardisierung von Untersuchungszugängen wünschenswert.

Practitioners’ perception of the impact of the vision of policy-makers on practice: The example of the recommendations of the Moser Commission

This paper presents the perception of practitioners of the impact of the Moser Committee recommendations and the Skills for Life agenda it generated. The paper further explores areas of convergence and divergence between practitioners’ perceptions and the underpinning values of the Moser Committee recommendations. The study utilised a range of research tools including an online questionnaire, documentary analysis and elements of discourse analysis in the collection and analysis of data. It found that there is substantial divergence between the perception of practitioners and the values underpinning policy. It concludes by suggesting that a varying perception of what constitutes sustainable education and the lack of input from practitioners into policy might be responsible for this significant divergence of opinion and also raised a question on the perceived role of practitioners in the policy‐making process.

English Channel towed sledge seabed images. Phase 1: scoping study and example analysis

During the 1970’s and 1980’s, the late Dr Norman Holme undertook extensive towed sledge surveys in the English Channel and some in the Irish Sea. Only a minority of the resulting images were analysed and reported before his death in 1989 but logbooks, video and film material has been archived in the National Marine Biological Library (NMBL) in Plymouth. A scoping study was therefore commissioned by the Joint Nature Conservation Committee and as a part of the Mapping European Seabed Habitats (MESH) project to identify the value of the material archived and the procedure and cost to undertake further work. The results of the scoping study are: 1. NMBL archives hold 106 videotapes (reel-to-reel Sony HD format) and 59 video cassettes (including 15 from the Irish Sea) in VHS format together with 90 rolls of 35 mm colour transparency film (various lengths up to about 240 frames per film). These are stored in the Archive Room, either in a storage cabinet or in original film canisters. 2. Reel-to-reel material is extensive and had already been selectively copied to VHS cassettes. The cost of transferring it to an accepted ‘long-life’ medium (Betamax) would be approximately £15,000. It was not possible to view the tapes as a suitable machine was not located. The value of the tapes is uncertain but they are likely to become beyond salvation within one to two years. 3. Video cassette material is in good condition and is expected to remain so for several more years at least. Images viewed were generally of poor quality and the speed of tow often makes pictures blurred. No immediate action is required. 4. Colour transparency films are in good condition and the images are very clear. They provide the best source of information for mapping seabed biotopes. They should be scanned to digital format but inexpensive fast copying is problematic as there are no between-frame breaks between images and machines need to centre the image based on between-frame breaks. The minimum cost to scan all of the images commercially is approximately £6,000 and could be as much as £40,000 on some quotations. There is a further cost in coding and databasing each image and, all-in-all it would seem most economic to purchase a ‘continuous film’ scanner and undertake the work in-house. 5. Positional information in ships logs has been matched to films and to video tapes. Decca Chain co-ordinates recorded in the logbooks have been converted to latitude and longitude (degrees, minutes and seconds) and a further routine developed to convert to degrees and decimal degrees required for GIS mapping. However, it is unclear whether corrections to Decca positions were applied at the time the position was noted. Tow tracks have been mapped onto an electronic copy of a Hydrographic Office chart. 6. The positions of start and end of each tow were entered to a spread sheet so that they can be displayed on GIS or on a Hydrographic Office Chart backdrop. The cost of the Hydrographic Office chart backdrop at a scale of 1:75,000 for the whole area was £458 incl. VAT. 7. Viewing all of the video cassettes to note habitats and biological communities, even by an experienced marine biologist, would take at least in the order of 200 hours and is not recommended. English Channel towed sledge seabed images. Phase 1: scoping study and example analysis. 6 8. Once colour transparencies are scanned and indexed, viewing to identify seabed habitats and biological communities would probably take about 100 hours for an experienced marine biologist and is recommended. 9. It is expected that identifying biotopes along approximately 1 km lengths of each tow would be feasible although uncertainties about Decca co-ordinate corrections and exact positions of images most likely gives a ±250 m position error. More work to locate each image accurately and solve the Decca correction question would improve accuracy of image location. 10. Using codings (produced by Holme to identify different seabed types), and some viewing of video and transparency material, 10 biotopes have been identified, although more would be added as a result of full analysis. 11. Using the data available from the Holme archive, it is possible to populate various fields within the Marine Recorder database. The overall ‘survey’ will be ‘English Channel towed video sled survey’. The ‘events’ become the 104 tows. Each tow could be described as four samples, i.e. the start and end of the tow and two areas in the middle to give examples along the length of the tow. These samples would have their own latitude/longitude co-ordinates. The four samples would link to a GIS map. 12. Stills and video clips together with text information could be incorporated into a multimedia presentation, to demonstrate the range of level seabed types found along a part of the northern English Channel. More recent images taken during SCUBA diving of reef habitats in the same area as the towed sledge surveys could be added to the Holme images.

How should sparse marine in situ measurements be compared to a continuous model: an example

This work demonstrates an example of the importance of an adequate method to sub-sample model results when comparing with in situ measurements. A test of model skill was performed by employing a point-to-point method to compare a multi-decadal hindcast against a sparse, unevenly distributed historic in situ dataset. The point-to-point method masked out all hindcast cells that did not have a corresponding in situ measurement in order to match each in situ measurement against its most similar cell from the model. The application of the point-to-point method showed that the model was successful at reproducing the inter-annual variability of the in situ datasets. Furthermore, this success was not immediately apparent when the measurements were aggregated to regional averages. Time series, data density and target diagrams were employed to illustrate the impact of switching from the regional average method to the point-to-point method. The comparison based on regional averages gave significantly different and sometimes contradicting results that could lead to erroneous conclusions on the model performance. Furthermore, the point-to-point technique is a more correct method to exploit sparse uneven in situ data while compensating for the variability of its sampling. We therefore recommend that researchers take into account for the limitations of the in situ datasets and process the model to resemble the data as much as possible.