Learning from humans: computational modeling of face recognition

In this paper, we propose a computational architecture of face recognition based on evidence from cognitive research. Several recent psychophysical experiments have shown that humans process faces by a combination of configural and component information. Using an appearance-based implementation of this architecture based on low-level features and their spatial relations, we were able to model aspects of human performance found in psychophysical studies. Furthermore, results from additional computational recognition experiments show that our framework is able to achieve excellent recognition performance even under large view rotations. Our interdisciplinary study is an example of how results from cognitive research can be used to construct recognition systems with increased performance. Finally, our modeling results also make new experimental predictions that will be tested in further psychophysical studies, thus effectively closing the loop between psychophysical experimentation and computational modeling.     

3D modeling of multiple-object scenes from sets of images

This paper proposes a new approach for multi-object 3D scene modeling. Scenes with multiple objects are characterized by object occlusions under several views, complex illumination conditions due to multiple reflections and shadows, as well as a variety of object shapes and surface properties. These factors raise huge challenges when attempting to model real 3D multi-object scene by using existing approaches which are designed mainly for single object modeling. The proposed method relies on the initialization provided by a rough 3D model of the scene estimated from the given set of multi-view images. The contributions described in this paper consists of two new methods for identifying and correcting errors in the reconstructed 3D scene. The first approach corrects the location of 3D patches from the scene after detecting the disparity between pairs of their projections into images. The second approach is called shape-from-contours and identifies discrepancies between projections of 3D objects and their corresponding contours, segmented from images. Both unsupervised and supervised segmentations are used to define the contours of objects.     

Identification of heat integration retrofit opportunities for crude distillation and residue cracking units

This study investigates improving the energy efficiency of two key refining processes: the Crude Distillation Unit (CDU) and the Residue Cracking Unit (RCU). The research methodology followed the ‘targeting before design’ approach. The CDU is a ‘tightly pinched’ system, with limited opportunities for further energy savings. The RCU actual ΔT_min is around 55 °C indicating a low level of current heat recovery. The Total-Site analysis shows that theoretically 18 MW of heat could be transferred from the RCU to the CDU, reducing CDU requirements by 40% for a new or grass roots design. RCU retrofit designs were developed to increase steam generation by up to 35% and in line with targeting estimates would appear to have economic potential. The alternative CDU-RCU retrofit design was developed to decrease CDU hot utility use. Although the Total-Site profile demonstrated strong potential for heat integration, this retrofit design is not commercially attractive, as the decrease in CDU fuel does not offset the cost of reduced steam generation. This demonstrates the need to consider the different fuel and steam costs in the Total-Site analysis.     

Critical Assessment 25: Friction stir processing,potential and problems

This assessment considers recent work on friction stir processing (FSP), which has been demonstrated to be an effective method for grain refinement and synthesis of new alloys and composites. The grain refinement is attributed to high strain rates leading to recrystallisation, while external cooling suppresses grain growth during cooling. The technique is capable of producing nanocrystalline alloys, and also able to disperse nanoparticles into alloys. The mechanical properties of processed materials agree with a combination of existing models for grain refinement, and precipitate reinforcement theory. Further improvements in the technique may help deal with severe tool wear during the FSP of composites, and reduce the complexity of composite fabrication using novel processing methods and tooling.     

Commercialization of Micro-fabrication of Antenna-Coupled Transition Edge Sensor Bolometer Detectors for Studies of the Cosmic Microwave Background

We report on the development of commercially fabricated multichroic antenna-coupled transition edge sensor (TES) bolometer arrays for cosmic microwave background (CMB) polarimetry experiments. CMB polarimetry experiments have deployed instruments in stages. Stage II experiments deployed with O(1000) detectors and reported successful detection of B-mode (divergence-free) polarization pattern in the CMB. Stage III experiments have recently started observing with O(10,000) detectors with wider frequency coverage. A concept for a stage IV experiment, CMB-S4, is emerging to make a definitive measurement of CMB polarization from the ground with O(400,000) detectors. The orders of magnitude increase in detector count for CMB-S4 require a new approach in detector fabrication to increase fabrication throughput and reduce the cost. We report on collaborative efforts with two commercial micro-fabrication foundries to fabricate antenna-coupled TES bolometer detectors. The detector design is based on the sinuous antenna-coupled dichroic detector from the POLARBEAR-2 experiment. The TES bolometers showed the expected I–V response, and the RF performance agrees with the simulation. We will discuss the motivation, design consideration, fabrication processes, test results, and how industrial detector fabrication could be a path to fabricate hundreds of detector wafers for future CMB polarimetry experiments.     

Electric and magnetic properties of ferromagnetic/piezoelectric bilayered composite

One of the most promising ways for the realization of multi-functional materials is the integration of oxides with different properties in artificial heterostructures. In this paper, a novel piezoelectric–ferromagnetic heterostructure consisting of 0.92Na_0.5Bi_0.5TiO₃–0.08BaTiO₃ (abbreviated as BNT–BT_0.08) and CoFe₂O₄ layers is fabricated on Si–Pt substrate, by sol–gel method coupled with spin-coating technique. The composite thin film shows only perovskite Bi_0.5Na_0.5TiO₃-like rhombohedral phase and CoFe₂O₄ cubic phase. The thickness of CoFe₂O₄ and BNT–BT_0.08 layers is ~?280 and?~?400 nm, respectively. BNT–BT_0.08/CoFe₂O₄ heterostructure thin film shows a saturation magnetization of 0.11 emu/g at 5 K and 0.07 emu/g at 295 K, dielectric constant of 235 at 1 kHz and tunability of 70% at 1 kHz and an electric field E?=?110 kV/cm. The results reveal that the investigated hybrid piezoelectric/ferromagnetic structure shows piezoelectric behavior, good ferroelectric and ferromagnetic properties. This bilayer composite can be used in miniature low-frequency magnetic sensor and piezoelectric sensor for biomedical domain.     

Variability of the midfacial muscles: analysis of 50 hemifacial cadaver dissections

The region of the midface represents a challenging area to both reconstructive and aesthetic surgeons. An anatomic study was performed that attempted to identify patterns and variations of the muscular anatomy. The goals of this study were twofold: to identify patterns and variability of the midfacial muscles that might impact on reconstructive efforts and to attempt to correlate this anatomy with features of the overlying soft tissues, specifically the nasolabial crease. Fifty hemifacial cadaver dissections were performed. The information collected was assembled into a large data base, and statistical significance was analyzed using Fisher's exact probability test. Results demonstrated that, although a great degree of variability exists with respect to the midfacial muscles, seven distinct patterns of these muscles did emerge. The most common pattern was the presence of a levator alae nasi, levator labii superioris, and zygomaticus major, which occurred in 44 percent of specimens. Specimens that possessed a risorius, zygomaticus minor, or both, were relatively uncommon. The consistent presence of the levators suggests adding a superior vector to recreate a smile in facial reanimation surgery. Two important anatomic variations were noted. A bifid zygomaticus major was found to be present in 34 percent of individuals. Because the inferior bundle had a dermocutaneous insertion, this anomaly may represent the anatomic correlate of a cheek "dimple." A second anomaly noted was the lateral cheek crease, which appeared to be associated with a cutaneous attachment from the underlying platysma muscle. However, no correlation could be found for facial muscle pattern and the overlying nasolabial crease structure. This lack of correlation may indicate that the facial muscles alone do not dictate the structure of the nasolabial crease and that other dynamic factors are involved in determining this feature of the aging face.     

Aligning with new digital strategy: A dynamic capabilities approach

Prior IS research has not fully addressed the aligning process in the highly dynamic context of digital strategy. To address this gap, we conduct a longitudinal analysis of a B2B company's journey to enact its B2C digital strategy, using the dynamic capabilities approach. We found that as an organization shifts towards a digital strategy, misalignments between the emergent strategy and resources give rise to tension. Our study resulted in the development of an aligning process model that is comprised of three phases (exploratory, building, and extending) and generalizable organizational aligning actions that form the organization's sensing, seizing, and transforming capacities. These aligning actions iteratively reconfigured organizational resources and refined strategy in order to respond to both changes in the environment and internal tensions. We also recognized that there are challenges to alignment, and conceptualized them as paradoxical tensions. This provided insights as to how such tensions are triggered and how they can be addressed. Finally, by applying the dynamic capabilities approach to aligning, we also show that alignment is not separate from such capabilities, but that aligning is enacted through the sensing, seizing and transforming capacities and their attendant aligning actions.     

An integrated semantics for reasoning about SysML design models using refinement

SysML is a variant of UML for systems design. Several formalisations of SysML (and UML) are available. Our work is distinctive in two ways: a semantics for refinement and for a representative collection of elements from the UML4SysML profile (blocks, state machines, activities, and interactions) used in combination. We provide a means to analyse and refine design models specified using SysML. This facilitates the discovery of problems earlier in the system development lifecycle, reducing time, and costs of production. Here, we describe our semantics, which is defined using a state-rich process algebra and implemented in a tool for automatic generation of formal models. We also show how the semantics can be used for refinement-based analysis and development. Our case study is a leadership-election protocol, a critical component of an industrial application. Our major contribution is a framework for reasoning using refinement about systems specified by collections of SysML diagrams.