10,000 publications from this institution
Object detection in cluttered, natural scenes has a high complexity since many local observations compete for object hypotheses. Voting methods provide an efficient solution to this problem. When Hough voting is extended to location and scale, votes naturally become lines through scale space due to the local scale-location-ambiguity. In contrast to this, current voting methods stick to the location-only setting and cast point votes, which require local estimates of scale. Rather than searching for object hypotheses in the Hough accumulator, we propose a weighted, pairwise clustering of voting lines to obtain globally consistent hypotheses directly. In essence, we propose a hierarchical approach that is based on a sparse representation of object boundary shape. Clustering of voting lines (CVL) condenses the information from these edge points in few, globally consistent candidate hypotheses. A final verification stage concludes by refining the candidates. Experiments on the ETHZ shape dataset show that clustering voting lines significantly improves state-of-the-art Hough voting techniques.
The mechanism for methanol oxidation on both TiO2 and V/TiO2 was investigated using temperature-programmed experiments with in-situ infrared spectroscopy. Infrared and Raman spectroscopy, along with XANES, show that the V/TiO2 sample consists predominantly of isolated VO4 units after calcination. Methanol was found to adsorb on the catalyst in three ways at 323 K: (1) molecularly, (2) across Ti−O−Ti bonds to form Ti−OCH3/Ti−OH pairs, and (3) across V−O−Ti bonds to form V−OCH3/Ti−OH pairs. Upon heating, two desorption peaks for CH3OH and H2O were observed on all samples below 500 K. Although TiO2 produced small amounts of CH2O, the addition of vanadium greatly enhanced the rate of formaldehyde formation. Also, on the V/TiO2 samples, it was noticed that the Ti−OCH3 groups disappear much more rapidly than on TiO2 alone. This is likely due to the reverse spillover of methoxide species from Ti to V, with the reaction occurring at lower temperatures at the vanadium center. Formate species were also detected during the experiments, and they are assumed to be intermediates in the decomposition of formaldehyde to CO, CO2, and H2O. The apparent activation energy of V/TiO2 for the formation of CH2O is 16 kcal/mol.
I'M GOING TO TALK about plug-and-play biology. And I'm g oing to first introduce it by talking about a problem, a problem many of you may know about. It's malaria. It actually affects about Ia quarter of the world's population; 2.4 billion people are at risk of getting malaria at any one time; between 300 and 500 million people have the disease, and every year between one and three million people die of the disease. Ninety percent are children under the age of five.And of those countries most affected by malaria, economists have noted that it reduces their GDP by roughly half. So it's a hugely debilitating disease. The drugs that are currently available are based largely on quinine. Quinine has been around for four hundred years; wars were fought for access to quinine. And it's been so widely used and abused that Plasmodium, which causes malaria, has grown resistant to it. So, widely available, inexpensive, and no longer effective.This is artemisinin (fig. 1). It comes from this plant Artemisia annua and it's a miracle cure, really, for malaria. It's been around for literally centuries. It was discovered in about 168 BC by the Chinese. There are some writings about its use, first for treating hemorrhoids and later for treating fevers. It was largely forgotten until the 1960s, when the Chinese were fighting in Vietnam and they needed a cure for the severe malaria there. They went back to their literature; they purified the active ingredient, and the rest is history.Now, here's how we get artemisinin. It's actually produced by this plant Artemisia annua in sacs called trichomes on the surface of the leaves. So there are basal cells in these trichomes and they push this oil, this artemisinin, up into these bags, and when you rub the leaf, you actually get it off.Now we get artemisinin from farmers who grow plants. There are a lot of plantations, primarily in Southeast Asia, where they grow the plant. They then purify it, from the plant; they purify it using the only solvent they have available-gasoline. There's a big problem. A few years ago when they were using leaded gasoline, the lead was contaminating the artemisinin supply, and then that's sold to the farmer for conversion into the various derivatives that are currently in use for treating malaria. Artesanate looks like it's going to be the best derivative of artemisinin because it's water soluble and it can be formulated in IVs.In 2005 the World Health Organization recommended artemisinin combination therapies as the drug of choice for treating malaria. These bars here (fig. 2) show artemisinin combination therapies delivered over that period. You can see there's a huge ramp-up primarily around 2006 and 2007. It takes about eighteen months to two years to get seeds in the field and artemisinin delivered to the farmers, so you need a very long lag time.Now here's where we project the demand will go. This is largely due to the Gates Foundation, the Clinton Global Fund, and the World Bank's making these drugs available. So there's going to be a huge demand for them. We're going to need on the order of 300 million treatments every year for malaria.Now when artemisinin was recommended as the drug of choice for treating malaria in 2005, the price spiked. This is one of the severe problems with artemisinin: the price fluctuates while there is an open market.So the price spiked and farmers started planting it like crazy, and that is shown right here (fig. 3). By 2007 there was a huge over-supply of artemisinin. What happened? They stopped planting it. The price of food went up, and they are no longer planting artemisinin. You can see that the prices are sitting down here because we have this over-supply, but the problem is that we're living on that over-supply from 2007.And guess what? Next year that supply runs out. And in fact we have a deficit. And if you look forward, there's going to be a huge u nder-supply by about 250 million treatments, which means the prices are going to go back up. …
While a number of gases are implicated in global warming, carbon dioxide is the most important contributor, and in one sense the entire phenomena can be seen as a human-induced perturbation of the carbon cycle. The Global Carbon Cycle offers a scientific assessment of the state of current knowledge of the carbon cycle by the world's leading scientists sponsored by SCOPE and the Global Carbon Project, and other international partners. It gives an introductory over-view of the carbon cycle, with multidisciplinary contributions covering biological, physical, and social science aspects. Included are 29 chapters covering topics including: an assessment of carbon-climate-human interactions; a portfolio of carbon management options; spatial and temporal distribution of sources and sinks of carbon dioxide; socio-economic driving forces of emissions scenarios. \n \nThroughout, contributors emphasize that all parts of the carbon cycle are interrelated, and only by developing a framework that considers the full set of feedbacks will we be able to achieve a thorough understanding and develop effective management strategies. \n \nThe Global Carbon Cycle edited by Christopher B. Field and Michael R. Raupach is part of the Rapid Assessment Publication series produced by the Scientific Committee on Problems of the Environment (SCOPE), in an effort to quickly disseminate the collective knowledge of the world's leading experts on topics of pressing environmental concern. \n- See more at: http://islandpress.org/book/the-global-carbon-cycle#sthash.CbWGvYvH.dpuf
Silicon photonics is a rapidly maturing platform for optical communication and sensing. As systems leveraging silicon photonics have grown in size and complexity, so too has the demand for high performance silicon photonics components. In order to meet these demands, inverse electromagnetic design has gained attention as a general tool for optimizing sophisticate silicon photonic devices. In this talk, we demonstrate how inverse electromagnetic design methods can be systematically applied to design high performance devices which both satisfy practical fabrication constraints and which are insensitive to manufacturing variations. These results pave the way for high efficiency silicon photonic component libraries.
The Covid-19 pandemic will have major implications for public trust in scientific expertise. But will this effect be positive or negative, and will it affect trust in individual scientists or science more broadly? Cevat Giray Aksoy, Barry Eichengreen and Orkun Saka write that people aged 18 to 25, whose core beliefs are still being formed, are likely to suffer the strongest impact of Covid-19 on their faith in scientists, but not in science.
