An advantage of D2C is that the counter can be learned reliably with additional synthesized probability maps. This addresses important data deficiency and sample imbalanced problems in counting. Our framework also enables easy diagnoses and analyses of error patterns. For instance, we find that, the counter per se is sufficiently accurate, while the bottleneck appears to be PMR. We further instantiate a network D2CNet in our framework and report state-of-the-art counting and localization performance across 6 crowd counting benchmarks. Since the probability map is a representation independent of visual appearance, D2CNet also exhibits remarkable cross-dataset transferability. Code and pretrained models are made available at https//git.io/d2cnet.This paper addresses the guided depth completion task in which the goal is to predict a dense depth map given a guidance RGB image and sparse depth measurements. Recent advances on this problem nurture hopes that one day we can acquire accurate and dense depth at a very low cost. A major challenge of guided depth completion is to effectively make use of extremely sparse measurements, e.g., measurements covering less than 1% of the image pixels. In this paper, we propose a fully differentiable model that avoids convolving on sparse tensors by jointly learning depth interpolation and refinement. More specifically, we propose a differentiable kernel regression layer that interpolates the sparse depth measurements via learned kernels. We further refine the interpolated depth map using a residual depth refinement layer which leads to improved performance compared to learning absolute depth prediction using a vanilla network. We provide experimental evidence that our differentiable kernel regression layer not only enables end-to-end training from very sparse measurements using standard convolutional network architectures, but also leads to better depth interpolation results compared to existing heuristically motivated methods. We demonstrate that our method outperforms many state-of-the-art guided depth completion techniques on both NYUv2 and KITTI. We further show the generalization ability of our method with respect to the density and spatial statistics of the sparse depth measurements.The lack of automatic tools to identify giant panda makes it hard to keep track of and manage giant pandas in wildlife conservation missions. In this paper, we introduce a new Giant Panda Identification (GPID) task, which aims to identify each individual panda based on an image. Though related to the human re-identification and animal classification problem, GPID is extraordinarily challenging due to subtle visual differences between pandas and cluttered global information. In this paper, we propose a new benchmark dataset iPanda-50 for GPID. The iPanda-50 consists of 6, 874 images from 50 giant panda individuals, and is collected from panda streaming videos. We also introduce a new Feature-Fusion Network with Patch Detector (FFN-PD) for GPID. The proposed FFN-PD exploits the patch detector to detect discriminative local patches without using any part annotations or extra location sub-networks, and builds a hierarchical representation by fusing both global and local features to enhance the inter-layer patch feature interactions. Specifically, an attentional cross-channel pooling is embedded in the proposed FFN-PD to improve the identify-specific patch detectors. Experiments performed on the iPanda-50 datasets demonstrate the proposed FFN-PD significantly outperforms competing methods. Besides, experiments on other fine-grained recognition datasets (i.e., CUB-200-2011, Stanford Cars, and FGVC-Aircraft) demonstrate that the proposed FFN-PD outperforms existing state-of-the-art methods.Recently, deep learning has drawn broad attention in the hyperspectral image (HSI) classification task. Many works have focused on elaborately designing various spectral-spatial networks, where convolutional neural network (CNN) is one of the most popular structures. To explore the spatial information for HSI classification, pixels with its adjacent pixels are usually directly cropped from hyperspectral data to form HSI cubes in CNN-based methods. However, the spatial land-cover distributions of cropped HSI cubes are usually complicated. The land-cover label of a cropped HSI cube cannot simply be determined by its center pixel. In addition, the spatial land-cover distribution of a cropped HSI cube is fixed and has less diversity. For CNN-based methods, training with cropped HSI cubes will result in poor generalization to the changes of spatial land-cover distributions. In this paper, an end-to-end fully convolutional segmentation network (FCSN) is proposed to simultaneously identify land-cover labels of all pixels in a HSI cube. First, several experiments are conducted to demonstrate that recent CNN-based methods show the weak generalization capabilities. Second, a fine label style is proposed to label all pixels of HSI cubes to provide detailed spatial land-cover distributions of HSI cubes. Third, a HSI cube generation method is proposed to generate plentiful HSI cubes with fine labels to improve the diversity of spatial land-cover distributions. Finally, a FCSN is proposed to explore spectral-spatial features from finely labeled HSI cubes for HSI classification. Experimental results show that FCSN has the superior generalization capability to the changes of spatial land-cover distributions.Batteryless, wireless, and packageless acoustic wave sensors are particularly desirable for harsh high-temperature environments. In this letter, an acoustic wave sensor based on a lithium niobate (Y + 128° cut, abbreviated LN-Y128) substrate with a buried platinum interdigital transducer (IDT) in an aluminum nitride (AlN) overlayer is investigated. Previously, it was demonstrated theoretically that due to the specific properties of LN-Y128, Rayleigh-type guided waves can propagate at the AlN/IDT(Pt)/LN-Y128 interface. Here, this structure is, for the first time, studied experimentally, including the growth and properties of the AlN layer onto irregular platinum IDTs. Both Shear Horizontal and Rayleigh-type waves have been identified after the AlN deposition and the velocities are consistent with the fitted SDA-FEM-SDA (a combination of finite element modeling with spectral domain analysis) simulations. Electrical measurements with a surface perturbation and temperature measurements show that the AlN/IDT(Pt)/LN-Y128 bilayer structure is promising as a packageless high-temperature sensor.Automated Machine Learning (AutoML) seeks to automatically find so-called machine learning pipelines that maximize the prediction performance when being used to train a model on a given dataset. One of the main and yet open challenges in AutoML is an effective use of computational resources An AutoML process involves the evaluation of many candidate pipelines, which are costly but often ineffective because they are canceled due to a timeout. In this paper, we present an approach to predict the runtime of two-step machine learning pipelines with up to one pre-processor, which can be used to anticipate whether or not a pipeline will time out. Separate runtime models are trained offline for each algorithm that may be used in a pipeline, and an overall prediction is derived from these models. We empirically show that the approach increases successful evaluations made by an AutoML tool while preserving or even improving on the previously best solutions.
To treat tissues that are difficult to access, ultrasound based minimally invasive treatment (MIT) is promising. However, high-power ultrasound delivery through waveguides had been difficult which can increase treatment duration. It is our effort to design the waveguide that can transmit powerful ultrasound.
The waveguide with two parabolic reflectors was proposed by us to produce high-energy-density plane wave. Use of flexible and long thin waveguide was demonstrated here.
Double Parabolic refLectors wave-guided high-power Ultrasonic tranSducer (DPLUS) including a ϕ1 mmx1 m Nitinol thin waveguide was fabricated. It was shown that high-power ultrasound between 1 to 2 MHz can be propagated through the thin waveguide. Low-loss waveguide material was confirmed to be important to enhance output. As ultrasound is transmitted into working medium, energy mainly flows from the side surface. Temperature of target soft tissue was demonstrated to drastically increase by 10 degree in 30 seconds.
The developed DPLUS makes high-power ultrasound transmission in long and flexible thin waveguide possible.
The concept of DPLUS for delivering high-power ultrasound is powerful in the field of Ultrasonics.
The concept of DPLUS for delivering high-power ultrasound is powerful in the field of Ultrasonics.
The Milan metropolitan area in Northern Italy was among the most severely hit by the SARS-CoV-2 outbreak. The aim of this study was to examine the seroprevalence trends of SARS-CoV-2 in healthy asymptomatic adults, and the risk factors and laboratory correlates of positive tests.
We conducted a cross-sectional study in a random sample of blood donors, who were asymptomatic at the time of evaluation, at the beginning of the first phase (February 24
to April 8
2020; n=789). Presence of IgM/IgG antibodies against the SARS-CoV-2-Nucleocapsid protein was assessed by a lateral flow immunoassay.
The test had a 100/98.3 sensitivity/specificity (n=32/120 positive/negative controls, respectively), and the IgG test was validated in a subset by an independent ELISA against the Spike protein (n=34, p<0.001). At the start of the outbreak, the overall adjusted seroprevalence of SARS-CoV-2 was 2.7% (95% CI 0.3-6%; p<0.0001 vs 120 historical controls). During the study period, characterised by a gradual implementation of social distancing measures, there was a progressive increase in the adjusted seroprevalence to 5.2% (95% CI 2.4-9.0; 4.5%, 95% CI 0.9-9.2% according to a Bayesian estimate) due to a rise in IgG reactivity to 5% (95% CI 2.8-8.2; p=0.004 for trend), but there was no increase in IgM
(p=not significant). https://www.selleckchem.com/ At multivariate logistic regression analysis, IgG reactivity was more frequent in younger individuals (p=0.043), while IgM reactivity was more frequent in individuals aged >45 years (p=0.002).
SARS-CoV-2 infection was already circulating in Milan at the start of the outbreak. The pattern of IgM/IgG reactivity was influenced by age IgM was more frequently detected in participants aged >45 years. By the end of April, 2.4-9.0% of healthy adults had evidence of seroconversion.
45 years. By the end of April, 2.4-9.0% of healthy adults had evidence of seroconversion.
As of publication, a total of 41 null alleles have been acknowledged by the International Society of Blood Transfusion (ISBT) to cause the rare Jk
phenotype, but none have been discovered in Austria thus far.
Two patients with anti-Jk3 were serologically identified by a positive antibody screening and typed as Jk(a-b-). The initial genotyping using an SSP-PCR method for the common 838A/G polymorphism indicated a JK*02/02, or JK*01/02 genotype, respectively. To find the disruptive mutations, Sanger sequencing was performed and results were compared to the reference sequence. The patient's antibodies were characterized with a monocyte monolayer assay (MMA) for their potential clinical significance.
Three novel null-mutations of the SLC14A1 gene were found in two patients. Patient 1 was homozygous for a 10bp deletion in exon 4 (c.157_166del on JK*02). Testing of her family members revealed Mendelian inheritance of the deletional allele. The other patient was compound heterozygous for two mutations one allele carrying a single base deletion in exon 4 (c.