The Unmanned Aerial Vehicle Benchmark: Object Detection,Tracking and Baseline

International Journal of Computer Vision - With the increasing popularity of Unmanned Aerial Vehicles (UAVs) in computer vision-related applications, intelligent UAV video analysis has recently...     

Micro- and macrostructural characterization of polyvinylpirrolidone rotary-spun fibers

The application of high-speed rotary spinning can offer a useful mean for either preparation of fibrous intermediate for conventional dosage forms or drug delivery systems. Polyvinylpyrrolidone (PVP) and poly(vinylpyrrolidone-vinylacetate) (PVP VA) micro- and nanofibers of different polymer concentrations and solvent ratios were prepared with a high-speed rotary spinning technique. In order to study the influence of parameters that enable successful fiber production from polymeric viscous solutions, a complex micro- and macrostructural screening method was implemented. The obtained fiber mats were subjected to detailed morphological analysis using scanning electron microscope (SEM), and rheological measurements while the microstructural changes of fiber samples, based on the free volume changes, was analyzed by positron annihilation lifetime spectroscopy (PALS) and compared with their mechanical characteristics. The plasticizing effect of water tracked by ortho-positronium lifetime changes in relation to the mechanical properties of fibers. A concentration range of polyvinylpyrrolidone solutions was defined for the preparation of fibers of optimum fiber morphology and mechanical properties. The method enabled fiber formulation of advantageous functionality-related properties for further formulation of solid dosage forms.     

Correction to Said, Sebe, and Todorov (2009).

Reports an error in "Structural resemblance to emotional expressions predicts evaluation of emotionally neutral faces" by Christopher P. Said, Nicu Sebe and Alexander Todorov (Emotion, 2009[Apr], Vol 9[2], 260-264). In this article a symbol was incorrectly omitted from Figure 1, part C. To see the complete article with the corrected figure, please go to http://dx.doi.org/10.1037/a0014681. (The following abstract of the original article appeared in record 2009-04472-011.) People make trait inferences based on facial appearance despite little evidence that these inferences accurately reflect personality. The authors tested the hypothesis that these inferences are driven in part by structural resemblance to emotional expressions. The authors first had participants judge emotionally neutral faces on a set of trait dimensions. The authors then submitted the face images to a Bayesian network classifier trained to detect emotional expressions. By using a classifier, the authors can show that neutral faces perceived to possess various personality traits contain objective resemblance to emotional expression. In general, neutral faces that are perceived to have positive valence resemble happiness, faces that are perceived to have negative valence resemble disgust and fear, and faces that are perceived to be threatening resemble anger. These results support the idea that trait inferences are in part the result of an overgeneralization of emotion recognition systems. Under this hypothesis, emotion recognition systems, which typically extract accurate information about a person's emotional state, are engaged during the perception of neutral faces that bear subtle resemblance to emotional expressions. These emotions could then be misattributed as traits. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Active domain adaptation with noisy labels for multimedia analysis

Supervised learning methods require sufficient labeled examples to learn a good model for classification or regression. However, available labeled data are insufficient in many applications. Active learning (AL) and domain adaptation (DA) are two strategies to minimize the required amount of labeled data for model training. AL requires the domain expert to label a small number of highly informative examples to facilitate classification, while DA involves tuning the source domain knowledge for classification on the target domain. In this paper, we demonstrate how AL can efficiently minimize the required amount of labeled data for DA. Since the source and target domains usually have different distributions, it is possible that the domain expert may not have sufficient knowledge to answer each query correctly. We exploit our active DA framework to handle incorrect labels provided by domain experts. Experiments with multimedia data demonstrate the efficiency of our proposed framework for active DA with noisy labels.     

Authentic facial expression analysis

There is a growing trend toward emotional intelligence in human–computer interaction paradigms. In order to react appropriately to a human, the computer would need to have some perception of the emotional state of the human. We assert that the most informative channel for machine perception of emotions is through facial expressions in video. One current difficulty in evaluating automatic emotion detection is that there are currently no international databases which are based on authentic emotions. The current facial expression databases contain facial expressions which are not naturally linked to the emotional state of the test subject. Our contributions in this work are twofold: first, we create the first authentic facial expression database where the test subjects are showing the natural facial expressions based upon their emotional state. Second, we evaluate the several promising machine learning algorithms for emotion detection which include techniques such as Bayesian networks, SVMs, and decision trees.     

Spin-Orbit Readout Using Thin Films of Topological Insulator Sb2Te3 Deposited by Industrial Magnetron Sputtering

Driving a spin-logic circuit requires the production of a large output signal by spin-charge interconversion in spin-orbit readout devices. This should be possible by using topological insulators, which are known for their high spin-charge interconversion efficiency. However, high-quality topological insulators have so far only been obtained on a small scale, or with large scale deposition techniques that are not compatible with conventional industrial deposition processes. The nanopatterning and electrical spin injection into these materials have also proven difficult due to their fragile structure and low spin conductance. The fabrication of a spin-orbit readout device from the topological insulator Sb₂Te₃ deposited by large-scale industrial magnetron sputtering on SiO₂ is presented. Despite a modification of the Sb₂Te₃ layer structural properties during the device nanofabrication, a sizeable output voltage is measured that can be unambiguously ascribed to a spin-charge interconversion process. The results pave the way for the integration of layered van der Waals materials in spin-logic devices.     

A 45-nm Bulk CMOS Embedded SRAM With Improved Immunity Against Process and Temperature Variations

The variation tolerant assist circuits of an SRAM against process and temperature are proposed. Passive resistances are introduced to the read assist circuit with replica memory transistors to lower the wordline voltage accurately reflecting the process and temperature variations. For the sake of not only enlarging the write margin but also reducing power consumption and speed overhead, the divided dynamic power-line scheme based on a charge sharing is adopted. Test chips of 512-Kb SRAM macros and isolated memory cell TEGs are fabricated using 45-nm bulk CMOS technology. Two types of 6-T SRAM cells, whose sizes were 0.245 mum² and 0.327 mum² were designed and evaluated. From the measurement results, we achieved over 100-mV improvement for static noise margin, and 35 mV for write margin for both SRAM cells at 1.0-V worst condition by using assist circuitry. It enables the wordline level to keep higher voltage at the slowest condition than the typical process condition, which results in 83% improvement of the cell current compared with the conventional assist circuit. Furthermore, the minimum operating voltage in the worst case condition was improved by 170 mV, confirming a high immunity against process and temperature variations with less than 10% area overhead.