Person Re-Identification with Improved Performance by Incorporating Focal Tversky Loss in AGW Baseline.

Person re-identification (re-ID) is one of the essential tasks for modern visual intelligent systems to identify a person from images or videos captured at different times, viewpoints, and spatial positions. In fact, it is easy to make an incorrect estimate for person re-ID in the presence of illumination change, low resolution, and pose differences. To provide a robust and accurate prediction, machine learning techniques are extensively used nowadays. However, learning-based approaches often face difficulties in data imbalance and distinguishing a person from others having strong appearance similarity. To improve the overall re-ID performance, false positives and false negatives should be part of the integral factors in the design of the loss function. In this work, we refine the well-known AGW baseline by incorporating a focal Tversky loss to address the data imbalance issue and facilitate the model to learn effectively from the hard examples. Experimental results show that the proposed re-ID method reaches rank-1 accuracy of 96.2% (with mAP: 94.5) and rank-1 accuracy of 93% (with mAP: 91.4) on Market1501 and DukeMTMC datasets, respectively, outperforming the state-of-the-art approaches.

RNA Interference-Mediated Knockdown of Genes Encoding Spore Wall Proteins Confers Protection against Nosema ceranae Infection in the European Honey Bee, Apis mellifera.

Nosema ceranae (Opisthosporidia: Microsporidia) is an emergent intracellular parasite of the European honey bee (Apis mellifera) and causes serious Nosema disease which has been associated with worldwide honey bee colony losses. The only registered treatment for Nosema disease is fumagillin-b, and this has raised concerns about resistance and off-target effects. Fumagillin-B is banned from use in honey bee colonies in many countries, particularly in Europe. As a result, there is an urgent need for new and effective therapeutic options to treat Nosema disease in honey bees. An RNA interference (RNAi)-based approach can be a potent strategy for controlling diseases in honey bees. We explored the therapeutic potential of silencing the sequences of two N. ceranae encoded spore wall protein (SWP) genes by means of the RNAi-based methodology. Our study revealed that the oral ingestion of dsRNAs corresponding to SWP8 and SWP12 used separately or in combination could lead to a significant reduction in spore load, improve immunity, and extend the lifespan of N. ceranae-infected bees. The results from the work completed here enhance our understanding of honey bee host responses to microsporidia infection and highlight that RNAi-based therapeutics are a promising treatment for honey bee diseases.

Deep Learning-Based Real-Time Multiple-Person Action Recognition System.

Action recognition has gained great attention in automatic video analysis, greatly reducing the cost of human resources for smart surveillance. Most methods, however, focus on the detection of only one action event for a single person in a well-segmented video, rather than the recognition of multiple actions performed by more than one person at the same time for an untrimmed video. In this paper, we propose a deep learning-based multiple-person action recognition system for use in various real-time smart surveillance applications. By capturing a video stream of the scene, the proposed system can detect and track multiple people appearing in the scene and subsequently recognize their actions. Thanks to high resolution of the video frames, we establish a zoom-in function to obtain more satisfactory action recognition results when people in the scene become too far from the camera. To further improve the accuracy, recognition results from inflated 3D ConvNet (I3D) with multiple sliding windows are processed by a nonmaximum suppression (NMS) approach to obtain a more robust decision. Experimental results show that the proposed method can perform multiple-person action recognition in real time suitable for applications such as long-term care environments.

Iterative Pose Refinement for Object Pose Estimation Based on RGBD Data.

Accurate estimation of 3D object pose is highly desirable in a wide range of applications, such as robotics and augmented reality. Although significant advancement has been made for pose estimation, there is room for further improvement. Recent pose estimation systems utilize an iterative refinement process to revise the predicted pose to obtain a better final output. However, such refinement process only takes account of geometric features for pose revision during the iteration. Motivated by this approach, this paper designs a novel iterative refinement process that deals with both color and geometric features for object pose refinement. Experiments show that the proposed method is able to reach 94.74% and 93.2% in ADD(-S) metric with only 2 iterations, outperforming the state-of-the-art methods on the LINEMOD and YCB-Video datasets, respectively.

Analysis of Disparity Error for Stereo Autofocus.

As more and more stereo cameras are installed on electronic devices, we are motivated to investigate how to leverage disparity information for autofocus. The main challenge is that stereo images captured for disparity estimation are subject to defocus blur unless the lenses of the stereo cameras are at the in-focus position. Therefore, it is important to investigate how the presence of defocus blur would affect stereo matching and, in turn, the performance of disparity estimation. In this paper, we give an analytical treatment of this fundamental issue of disparity-based autofocus by examining the relation between image sharpness and disparity error. A statistical approach that treats the disparity estimate as a random variable is developed. Our analysis provides a theoretical backbone for the empirical observation that, regardless of the initial lens position, disparity-based autofocus can bring the lens to the hill zone of the focus profile in one movement. The insight gained from the analysis is useful for the implementation of an autofocus system.

New evidence showing that the destruction of gut bacteria by antibiotic treatment could increase the honey bee's vulnerability to Nosema infection.

It has become increasingly clear that gut bacteria play vital roles in the development, nutrition, immunity, and overall fitness of their eukaryotic hosts. We conducted the present study to investigate the effects of gut microbiota disruption on the honey bee's immune responses to infection by the microsporidian parasite Nosema ceranae. Newly emerged adult workers were collected and divided into four groups: Group I-no treatment; Group II-inoculated with N. ceranae, Group III-antibiotic treatment, and Group IV-antibiotic treatment after inoculation with N. ceranae. Our study showed that Nosema infection did not cause obvious disruption of the gut bacterial community as there was no significant difference in the density and composition of gut bacteria between Group I and Group II. However, the elimination of gut bacteria by antibiotic (Groups III and IV) negatively impacted the functioning of the honey bees' immune system as evidenced by the expression of genes encoding antimicrobial peptides abaecin, defensin1, and hymenoptaecin that showed the following ranking: Group I > Group II > Group III > Group IV. In addition, significantly higher Nosema levels were observed in Group IV than in Group II, suggesting that eliminating gut bacteria weakened immune function and made honey bees more susceptible to Nosema infection. Based on Group IV having displayed the highest mortality rate among the four experimental groups indicates that antibiotic treatment in combination with stress, associated with Nosema infection, significantly and negatively impacts honey bee survival. The present study adds new evidence that antibiotic treatment not only leads to the complex problem of antibiotic resistance but can impact honey bee disease resistance. Further studies aimed at specific components of the gut bacterial community will provide new insights into the roles of specific bacteria and possibly new approaches to improving bee health.

Evidence of the synergistic interaction of honey bee pathogens Nosema ceranae and Deformed wing virus.

Nosema ceranae and Deformed wing virus (DWV) are two of the most prevalent pathogens currently attacking Western honey bees, Apis mellifera, and often simultaneously infect the same hosts. Here we investigated the effect of N. ceranae and Deformed wing virus (DWV) interactions on infected honey bees under lab conditions and at different nutrition statuses. Our results showed that Nosema could accelerate DWV replication in infected bees in a dose-dependent manner at the early stages of DWV infection. When bees were restricted from pollen nutrition, inoculation with 1×10(4) and 1×10(5) spores/bee could cause a significant increase in DWV titer, while inoculation with 1×10(3) spores/bee did not show any significant effect on the DWV titer. When bees were provided with pollen, only inoculation with 1×10(5) spores/bee showed significant effect on DWV titer. However, our results also showed that the two pathogens did not act synergistically when the titer of DWV reached a plateau. This study suggests that the synergistic effect of N. ceranae and DWV is dosage- and nutrition-dependent and that the synergistic interactions between the two pathogens could have implications on honey bee colony losses.

Israeli acute paralysis virus: epidemiology, pathogenesis and implications for honey bee health.

Israeli acute paralysis virus (IAPV) is a widespread RNA virus of honey bees that has been linked with colony losses. Here we describe the transmission, prevalence, and genetic traits of this virus, along with host transcriptional responses to infections. Further, we present RNAi-based strategies for limiting an important mechanism used by IAPV to subvert host defenses. Our study shows that IAPV is established as a persistent infection in honey bee populations, likely enabled by both horizontal and vertical transmission pathways. The phenotypic differences in pathology among different strains of IAPV found globally may be due to high levels of standing genetic variation. Microarray profiles of host responses to IAPV infection revealed that mitochondrial function is the most significantly affected biological process, suggesting that viral infection causes significant disturbance in energy-related host processes. The expression of genes involved in immune pathways in adult bees indicates that IAPV infection triggers active immune responses. The evidence that silencing an IAPV-encoded putative suppressor of RNAi reduces IAPV replication suggests a functional assignment for a particular genomic region of IAPV and closely related viruses from the Family Dicistroviridae, and indicates a novel therapeutic strategy for limiting multiple honey bee viruses simultaneously and reducing colony losses due to viral diseases. We believe that the knowledge and insights gained from this study will provide a new platform for continuing studies of the IAPV-host interactions and have positive implications for disease management that will lead to mitigation of escalating honey bee colony losses worldwide.

Characterization of the pharmaceutical effect of drugs on atherosclerotic lesions in vivo using integrated fluorescence imaging and Raman spectral measurements.

Direct assessment of the vascular lesions of model animals in vivo is important for the development of new antiatherosclerotic drugs. Nevertheless, biochemical analysis of the lipid profile in blood in vitro remains the most common way to evaluate the therapeutic effect of drugs targeting atherosclerosis because of an inherent difficulty to access the vascular wall. Using hypercholesterolemic zebrafish, we present an orchestrated application of Raman spectral measurements and confocal fluorescence imaging to interrogate the pharmacological response of atherosclerotic lesions in situ and in vivo. For demonstration, we investigated two commonly prescribed antihyperlipidemic drugs, ezetimibe and atorvastatin. The treatment of ezetimibe or atorvastatin alone decreased effectively the deposition of lipids in the vascular wall, and a combined dose showed a synergistic effect. Atorvastatin exerted a profound antioxidative effect on vascular fatty lesions. Analysis of individual lesions shows further that these lesions exhibited a heterogeneous response to the treatment of atorvastatin; a significant fraction of, but not all, the lesions became nonoxidized after the intervention. Beyond its efficacies in suppressing both the accumulation and oxidation of vascular lipids, atorvastatin expedited the clearance of vascular lipids. The possession of pleotropic (multiple) therapeutic effects on vascular fatty lesions of hypercholesterolemic zebrafish by atorvastatin is notably consistent with the known pharmaceutical effects of this drug on human beings. These results improve our understanding of the antiatherosclerotic effect of drugs. We envisage that our approach has the potential to become a platform to predict the pharmaceutical effects of new drugs aiming to cure human atherosclerotic diseases.

Evaluation of cage designs and feeding regimes for honey bee (Hymenoptera: Apidae) laboratory experiments.

The aim of this study was to improve cage systems for maintaining adult honey bee (Apis mellifera L.) workers under in vitro laboratory conditions. To achieve this goal, we experimentally evaluated the impact of different cages, developed by scientists of the international research network COLOSS (Prevention of honey bee COlony LOSSes), on the physiology and survival of honey bees. We identified three cages that promoted good survival of honey bees. The bees from cages that exhibited greater survival had relatively lower titers of deformed wing virus, suggesting that deformed wing virus is a significant marker reflecting stress level and health status of the host. We also determined that a leak- and drip-proof feeder was an integral part of a cage system and a feeder modified from a 20-ml plastic syringe displayed the best result in providing steady food supply to bees. Finally, we also demonstrated that the addition of protein to the bees' diet could significantly increase the level ofvitellogenin gene expression and improve bees' survival. This international collaborative study represents a critical step toward improvement of cage designs and feeding regimes for honey bee laboratory experiments.

Spore Loads May Not be Used Alone as a Direct Indicator of the Severity of Nosema ceranae Infection in Honey Bees Apis mellifera (Hymenoptera:Apidae).

Nosema ceranae Fries et al., 1996, a microsporidian parasite recently transferred from Asian honey bees Apis cerana F., 1793, to European honey bees Apis mellifera L., 1758, has been suspected as one of the major culprits of the worldwide honey bee colony losses. Spore load is a commonly used criterion to describe the intensity of Nosema infection. In this study, by providing Nosema-infected bees with sterilized pollen, we confirmed that pollen feeding increased the spore loads of honey bees by several times either in the presence or absence of a queen. By changing the amount of pollen consumed by bees in cages, we showed that spore loads increased with an increase in pollen consumption. Nosema infections decrease honey bee longevity and transcription of vitellogenin, either with or without pollen feeding. However, the reduction of pollen consumption had a greater impact on honey bee longevity and vitellogenin level than the increase of spore counts caused by pollen feeding. These results indicate that spore loads may not be used alone as a direct indicator of the severity of N. ceranae infection in honey bees.