Search

Automated three-dimensional (3D) object reconstruction is the task of building a geometric representation of a physical object by means of sensing its surface. Even though new single-view reconstruction techniques can predict the surface, they lead to incomplete models, specially, for non-commons objects such as antique objects or art sculptures. Therefore, to achieve the task's goals, it is essential to automatically determine the locations where the sensor will be placed so that the surface will be completely observed. This problem is known as the next-best-view problem. In this paper, we propose a data-driven approach to address the problem. The proposed approach trains a 3D convolutional neural network (3D CNN) with previous reconstructions in order to regress the position of the next-best-view. To the best of our knowledge, this is one of the first works that directly infers the next-best-view in a continuous space using a data-driven approach for the 3D object reconstruction task. We have validated the proposed approach making use of two groups of experiments. In the first group, several variants of the proposed architecture are analyzed. Predicted next-best-views were observed to be closely positioned to the ground truth. In the second group of experiments, the proposed approach is requested to reconstruct several unseen objects, namely, objects not considered by the 3D CNN during training nor validation. Coverage percentages of up to 90 % were observed. With respect to current state-of-the-art methods, the proposed approach improves the performance of previous next-best-view classification approaches and it is quite fast in running time (3 frames per second), given that it does not compute the expensive ray tracing required by previous information metrics. [ABSTRACT FROM AUTHOR]

• A next-best-view approach based on supervised deep learning is proposed. • It is shown that the next-best-view problem can be addressed as a classification task. • The proposed approach is 1750 times faster than state-of-the-art methods. • An original three-dimensional convolutional neural network that learns to predict the next-best-view is introduced. Motivated by the advances in 3D sensing technology and the spreading of low-cost robotic platforms, 3D object reconstruction has become a common task in many areas. Nevertheless, the selection of the optimal sensor pose that maximizes the reconstructed surface is a problem that remains open. It is known in the literature as the next-best-view planning problem. In this paper, we propose a novel next-best-view planning scheme based on supervised deep learning. The scheme contains an algorithm for automatic generation of datasets and an original three-dimensional convolutional neural network (3D-CNN) used to learn the next-best-view. Unlike previous work where the problem is addressed as a search, the trained 3D-CNN directly predicts the sensor pose. We present an experimental comparison of the proposed architecture against two alternative networks; we also compare it with state-of-the-art next-best-view methods in the reconstruction of several unknown objects. Our method is faster and reaches high coverage. [ABSTRACT FROM AUTHOR]

Experimental melting studies were conducted on a nepheline mugearite composition to pressures of 31 kbar in the presence of 0–30% added water. A temperature maximum in the near-liquidus stability of amphibole (with olivine) was found for a water content of 3·5 wt % at a pressure of 14 kbar. This is interpreted to have petrogenetic significance for the derivation of nepheline mugearite magmas from nepheline hawaiite by amphibole-dominated fractional crystallization at depth within the lithospheric mantle. Synthetic liquids at progressively lower temperatures range to nepheline benmoreite compositions very similar to those of natural xenolith-bearing high-pressure lavas elsewhere, and support the hypothesis that continued fractional crystallization could lead to high-pressure phonolite liquids. Independent experimental data for a basanite composition modeled on a lava from the same igneous province (the Newer Basalts of Victoria) permit the inference that primary asthenospheric basanite magmas undergo polybaric fractional crystallization during ascent, and may evolve to liquids ranging from nepheline hawaiite to phonolite upon encountering cooler lithospheric mantle at depths of 42–50 km. Such a model is consistent with the presence in some evolved alkalic lavas of both lithospheric peridotite xenoliths indicative of similar depths and of megacryst suites that probably represent disrupted pegmatitic segregations precipitated from precursor alkalic magmas in conduit systems within lithospheric mantle. [ABSTRACT FROM AUTHOR]

Introduction: Allelic imbalance is a common genetic event in many types of malignancy. The recent introduction of high-density SNPA for the concurrent analysis of loss of heterozygosity (LOH) and changes in chromosome copy number (CN) in small amounts of DNA has facilitated a comprehensive, genome-wide analysis of tumour cell populations. This method is of value in malignancies where standard cytogenetic analysis is difficult to perform. Objective: To characterise LOH and CN alterations in leukaemic samples using SNPA. Patients and method: A cohort of 86 patients presenting with childhood ALL within the Northern Region of the UK (Sept 1986-Jan 2005) were analysed using GeneChip® Human Mapping 10K Array (Affymetrix, Ltd.). Good quality of DNA (250ng) was sent to MRC Geneservice (Cambridge) for sample analysis. Results from SNPA analysis were analysed using proprietary software (Affymetrix). The results spreadsheets obtained from MRC contained information about the identity of each SNP and its chromosomal location were exported to Excel spreadsheets for in house analysis. Results: Sixty-nine of the 86 samples (80%) showed one or more significant areas of LOH. Consideration of the CN of the regions affected suggested that this was of 2 types: LOH associated with CN reduction (deletion) and LOH associated with no CN change or copy neutral-LOH (acquired isodisomy, AID). Allelic imbalances have been frequently identified on chromosome 9p, 12p and 6q. Loss of 9p has been associated with tumourigenesis and to be progressive in some cases at relapse. A tumor suppressor gene, p16INK4a is always included in 9p deleted region and has been suggested to be involved in leukaemogenesis. Discussion and conclusion: This study indicated that application of SNPA is very useful to characterize allelic imbalance in childhood ALL. In addition, unlike other whole-genome screening methods, it can readily detect LOH associated with the preservation of the normal CN (AID). [ABSTRACT FROM AUTHOR]

: Background and Aims Parasitism is well understood to have wide-ranging deleterious effects on host performance in species thus far characterized. Photosynthetic performance reductions have been noted in the Striga–Zea mays association; however, no such information exists for facultative hemiparasitic plants and their hosts, nor are the effects of host species understood. : Methods Chlorophyll fluorimetry was used to study the effects of parasitism by the hemiparasite Rhinanthus minor on the grass Phleum bertolinii and the forb Plantago lanceolata, and the effects of host species on the photosynthetic apparatus of R. minor. : Key Results Parasitism by Rhinanthus led to a significant decrease in the host, and total (host + parasite) biomass in Phleum; however, in Plantago, no significant repression of growth was noted. Maximum quantum yield (Fv/Fm) was reduced in parasitized Plantago, relative to control plants, but not in Phleum. Fv/Fm was significantly lower in R. minor parasitizing Phleum than Plantago, suggesting Phleum to be a superior host to Plantago for R. minor. Steady-state quantum yield (ΦPSII) was significantly depressed in parasitized Phleum, but only at low irradiances in Plantago. ΦPSII was very low for R. minor grown on Plantago, but not Phleum. : Conclusions Shown here is the first evidence of the suppression of host photosynthesis by a facultative hemiparasitic plant, which has significant effects on total biomass production. Host identity is a significant factor in parasite success, with the forb Plantago lanceolata exhibiting apparent chemical as well as previously identified physical defences to parasitism. It is proposed that the electron transport rate (as denoted by ΦPSII) represents the limiting factor for biomass accumulation in this system, and that Plantago is able to suppress the growth of Rhinanthus by suppressing the electron transport rate. [ABSTRACT FROM AUTHOR]