High-Performance Azo Cathodes Enabled by N-Heteroatomic Substitution for Zinc Batteries with a Self-Charging Capability.

Redox-active azo compounds are emerging as promising cathode materials due to their multi-electron redox capacity and fast redox response. However, their practical application is often limited by low output voltage and poor thermal stability. Herein, we use a heteroatomic substitution strategy to develop 4,4'-azopyridine. This modification results in a 350 mV increase in reduction potential compared to traditional azobenzene, increasing the energy density at the material level from 187 to 291 Wh kg-1. The introduced heteroatoms not only raise the melting point of azo compounds from 68 °C to 112 °C by forming an intermolecular hydrogen-bond network but also improves electrode kinetics by reducing energy band gaps. Moreover, 4,4'-azopyridine forms metal-ligand complexes with Zn2+ ions, which further self-assemble into a robust superstructure, acting as a molecular conductor to facilitate charge transfer. Consequently, the batteries display a good rate performance (192 mAh g-1 at 20 C) and an ultra-long lifespan of 60,000 cycles. Notably, we disclose that the depleted batteries spontaneously self-charge when exposed to air, marking a significant advancement in the development of self-powered aqueous systems.

Multiview Deep Subspace Clustering Networks.

Multiview subspace clustering aims to discover the inherent structure of data by fusing multiple views of complementary information. Most existing methods first extract multiple types of handcrafted features and then learn a joint affinity matrix for clustering. The disadvantage of this approach lies in two aspects: 1) multiview relations are not embedded into feature learning and 2) the end-to-end learning manner of deep learning is not suitable for multiview clustering. Even when deep features have been extracted, it is a nontrivial problem to choose a proper backbone for clustering on different datasets. To address these issues, we propose the multiview deep subspace clustering networks (MvDSCNs), which learns a multiview self-representation matrix in an end-to-end manner. The MvDSCN consists of two subnetworks, i.e., a diversity network (Dnet) and a universality network (Unet). A latent space is built using deep convolutional autoencoders, and a self-representation matrix is learned in the latent space using a fully connected layer. Dnet learns view-specific self-representation matrices, whereas Unet learns a common self-representation matrix for all views. To exploit the complementarity of multiview representations, the Hilbert-Schmidt independence criterion (HSIC) is introduced as a diversity regularizer that captures the nonlinear, high-order interview relations. Because different views share the same label space, the self-representation matrices of each view are aligned to the common one by universality regularization. The MvDSCN also unifies multiple backbones to boost clustering performance and avoid the need for model selection. Experiments demonstrate the superiority of the MvDSCN.

Mesoporous PdBi nanocages for enhanced electrocatalytic performances by all-direction accessibility and steric site activation.

An effective yet simple approach was developed to synthesize mesoporous PdBi nanocages for electrochemical applications. This technique relies on the subtle utilization of the hydrolysis of a metal salt to generate precipitate cores in situ as templates for navigating the growth of mesoporous shells with the assistance of polymeric micelles. The mesoporous PdBi nanocages are then obtained by excavating vulnerable cores and regulating the crystals of mesoporous metallic skeletons. The resultant mesoporous PdBi nanocages exhibited excellent electrocatalytic performance toward the ethanol oxidation reaction with a mass activity of 3.56 A mg-1_Pd, specific activity of 17.82 mA cm-2 and faradaic efficiency of up to 55.69% for C1 products.

Shipborne expendable all-optical fiber ocean temperature-depth profile sensor.

Traditional electrical expendable bathythermograph (XBT) is designed to fall at a known rate based on a great deal of experiments so that the depth of the temperature profile can be inferred from the time it enters the water. Unlike the traditional electrical XBT, which derives the depth from fall-rate equations, we propose an all-optical fiber (AOF) XBT (AOF-XBT) based on cascade of two fiber Bragg gratings (FBGs). In the AOF-XBT, the depth data comes from one FBG, which responds in real time to the pressure acting on the diaphragm, and temperature data can be measured via the other FBG simultaneously. First, the pressure and temperature response characteristics of the AOF-XBT are analyzed based on a finite element method. Then, the temperature and pressure calibrations for the AOF-XBT is completed after they are packaged. Results show that the mean-temperature sensitivity of two sensors are 14.765 and 13.705 pm/°C in the range of 5°C-30°C, and the mean-pressure sensitivities are -2.75586 and -3.00472nm/MPa in the range of 0-0.6 MPa, respectively. At last, by comparing the results obtained from the AOF-XBT and the SBE 911plus CTD that tested in the sea area of Weihai, the trends of the temperature-depth profile from the two devices are consistent, which presents a new all-optical technique to provide full ocean temperature-depth profile observations.

Detection and Tracking Meet Drones Challenge.

Drones, or general UAVs, equipped with cameras have been fast deployed with a wide range of applications, including agriculture, aerial photography, and surveillance. Consequently, automatic understanding of visual data collected from drones becomes highly demanding, bringing computer vision and drones more and more closely. To promote and track the developments of object detection and tracking algorithms, we have organized three challenge workshops in conjunction with ECCV 2018, ICCV 2019 and ECCV 2020, attracting more than 100 teams around the world. We provide a large-scale drone captured dataset, VisDrone, which includes four tracks, i.e., (1) image object detection, (2) video object detection, (3) single object tracking, and (4) multi-object tracking. In this paper, we first present a thorough review of object detection and tracking datasets and benchmarks, and discuss the challenges of collecting large-scale drone-based object detection and tracking datasets with fully manual annotations. After that, we describe our VisDrone dataset, which is captured over various urban/suburban areas of 14 different cities across China from North to South. Being the largest such dataset ever published, VisDrone enables extensive evaluation and investigation of visual analysis algorithms for the drone platform. We provide a detailed analysis of the current state of the field of large-scale object detection and tracking on drones, and conclude the challenge as well as propose future directions. We expect the benchmark largely boost the research and development in video analysis on drone platforms. All the datasets and experimental results can be downloaded from https://github.com/VisDrone/VisDrone-Dataset.

Application of Variational Method to Stability Analysis of Cantilever Vertical Plates with Bimodular Effect.

In the design of cantilevered balconies of buildings, many stability problems exist concerning vertical plates, in which reaching a critical load plays an important role during the stability analysis of the plate. At the same time, the concrete forming vertical plate, as a typical brittle material, has larger compressive strength but lower tensile strength, which means the tensile and compression properties of concrete are different. However, due to the complexities of such analyses, this difference has not been considered. In this study, the variational method is used to analyze stability problems of cantilever vertical plates with bimodular effect, in which different loading conditions and plate shapes are also taken into account. For the effective implementation of a variational method, the bending strain energy based on bimodular theory is established first, and critical loads of four stability problems are obtained. The results indicate that the bimodular effect, as well as different loading types and plate shapes, have influences on the final critical loads, resulting in varying degrees of buckling. In particular, if the average value of the tensile modulus and compressive modulus remain unchanged, the introduction of the bimodular effect will weaken, to some extent, the bending stiffness of the plate. Among the four stability problems, a rectangular plate with its top and bottom loaded is most likely to buckle; next is a rectangular plate with its top loaded, followed by a triangular plate with its bottom loaded. A rectangular plate with its bottom loaded is least likely to buckle. This work may serve as a theoretical reference for the refined analysis of vertical plates. Plates are made of concrete or similar material whose bimodular effect is relatively obvious and cannot be ignored arbitrarily; otherwise the greater inaccuracies will be encountered in building designs.

Ultrasensitive temperature sensor and mode converter based on a modal interferometer in a two-mode fiber.

This paper presents an ultrasensitive temperature sensor and tunable mode converter based on an isopropanol-sealed modal interferometer in a two-mode fiber. The modal interferometer consists of a tapered two-mode fiber (TTMF) sandwiched between two single-mode fibers. The sensor provides high-sensitivity temperature sensing by taking advantages of TTMF, isopropanol and the Vernier-like effect. The TTMF provides a uniform modal interferometer with LP01 and LP11 modes as well as strong evanescent field on its surface. The temperature sensitivity of the sensor can be improved due to the high thermo-optic coefficient of isopropanol. The Vernier-like effect based on the overlap of two interference spectra is applied to magnify the sensing capabilities with a sensitivity magnification factor of 58.5. The temperature sensor is implemented by inserting the modal interferometer into an isopropanol-sealed capillary. The experimental and calculated results show the transmission spectrum exhibit blue shift with increasing ambient temperature. Experimental results show that the isopropanol-sealed modal interferometer provides a temperature sensitivity up to -140.5 nm/°C. The interference spectrum has multiple dips at which the input LP01 mode is converted to the LP11 mode. This modal interferometer acts as a tunable multi-channel mode converter. The mode converter that can be tuned by varying temperature and mode switch is realized.

Broadband mode-selective couplers based on tapered side-polished fibers.

We propose the broadband mode-selective coupler (MSC) formed with a side-polished six mode fiber (6MF) and a tapered side-polished small core single-mode fiber (SC-SMF) or an SMF. The MSCs are designed to allow the LP01 mode in the SC-SMF and SMF to completely couple to the LP01, LP11, LP21, LP02, LP31, LP12 modes in the 6MF over a broadband wavelength range. The phase-matching conditions of the MSCs are satisfied by tapering the SC-SMF and SMF to specific diameters. The tapered fibers are side-polished to designed residual fiber thickness using the wheel polishing technique. The effective indices of the side-polished fibers are measured with the prism coupling method. The MSCs provide high coupling ratio and high mode purity. High coupling efficiencies in excess of 81% for all the higher-order modes are obtained in the wavelength range 1530-1600 nm. For the LP01, LP11, LP21, LP02, LP31, LP12 MSCs at 1550 nm, the coupling ratios are 96.2%, 99.8%, 89.5%, 85.0%, 90.9%, 96.1%, respectively, and the mode purity of the MSCs is higher than 88.0%. The loss of the MSCs is lower than 1.8 dB in the wavelength range 1530-1600 nm. This device can be applied in broadband mode-division multiplexing transmission systems.

Graph Regularized Flow Attention Network for Video Animal Counting From Drones.

In this paper, we propose a large-scale video based animal counting dataset collected by drones (AnimalDrone) for agriculture and wildlife protection. The dataset consists of two subsets, i.e., PartA captured on site by drones and PartB collected from the Internet, with rich annotations of more than 4 million objects in 53, 644 frames and corresponding attributes in terms of density, altitude and view. Moreover, we develop a new graph regularized flow attention network (GFAN) to perform density map estimation in dense crowds of video clips with arbitrary crowd density, perspective, and flight altitude. Specifically, our GFAN method leverages optical flow to warp the multi-scale feature maps in sequential frames to exploit the temporal relations, and then combines the enhanced features to predict the density maps. Moreover, we introduce the multi-granularity loss function including pixel-wise density loss and region-wise count loss to enforce the network to concentrate on discriminative features for different scales of objects. Meanwhile, the graph regularizer is imposed on the density maps of multiple consecutive frames to maintain temporal coherency. Extensive experiments are conducted to demonstrate the effectiveness of the proposed method, compared with several state-of-the-art counting algorithms. The AnimalDrone dataset is available at https://github.com/VisDrone/AnimalDrone.

SiamCAN: Real-Time Visual Tracking Based on Siamese Center-Aware Network.

In this article, we present a novel Siamese center-aware network (SiamCAN) for visual tracking, which consists of the Siamese feature extraction subnetwork, followed by the classification, regression, and localization branches in parallel. The classification branch is used to distinguish the target from background, and the regression branch is introduced to regress the bounding box of the target. To reduce the impact of manually designed anchor boxes to adapt to different target motion patterns, we design the localization branch to localize the target center directly to assist the regression branch generating accurate results. Meanwhile, we introduce the global context module into the localization branch to capture long-range dependencies for more robustness to large displacements of the target. A multi-scale learnable attention module is used to guide these three branches to exploit discriminative features for better performance. Extensive experiments on 9 challenging benchmarks, namely VOT2016, VOT2018, VOT2019, OTB100, LTB35, LaSOT, TC128, UAV123 and VisDrone-SOT2019 demonstrate that SiamCAN achieves leading accuracy with high efficiency. Our source code is available at https://isrc.iscas.ac.cn/gitlab/research/siamcan.

Embedding Perspective Analysis Into Multi-Column Convolutional Neural Network for Crowd Counting.

The crowd counting is challenging for deep networks due to several factors. For instance, the networks can not efficiently analyze the perspective information of arbitrary scenes, and they are naturally inefficient to handle the scale variations. In this work, we deliver a simple yet efficient multi-column network, which integrates the perspective analysis method with the counting network. The proposed method explicitly excavates the perspective information and drives the counting network to analyze the scenes. More concretely, we explore the perspective information from the estimated density maps and quantify the perspective space into several separate scenes. We then embed the perspective analysis into the multi-column framework with a recurrent connection. Therefore, the proposed network matches various scales with the different receptive fields efficiently. Secondly, we share the parameters of the branches with various receptive fields. This strategy drives the convolutional kernels to be sensitive to the instances with various scales. Furthermore, to improve the evaluation accuracy of the column with a large receptive field, we propose a transform dilated convolution. The transform dilated convolution breaks the fixed sampling structure of the deep network. Moreover, it needs no extra parameters and training, and the offsets are constrained in a local region, which is designed for the congested scenes. The proposed method achieves state-of-the-art performance on five datasets (ShanghaiTech, UCF CC 50, WorldEXPO'10, UCSD, and TRANCOS).

Dual Encoding for Abstractive Text Summarization.

Recurrent neural network-based sequence-to-sequence attentional models have proven effective in abstractive text summarization. In this paper, we model abstractive text summarization using a dual encoding model. Different from the previous works only using a single encoder, the proposed method employs a dual encoder including the primary and the secondary encoders. Specifically, the primary encoder conducts coarse encoding in a regular way, while the secondary encoder models the importance of words and generates more fine encoding based on the input raw text and the previously generated output text summarization. The two level encodings are combined and fed into the decoder to generate more diverse summary that can decrease repetition phenomenon for long sequence generation. The experimental results on two challenging datasets (i.e., CNN/DailyMail and DUC 2004) demonstrate that our dual encoding model performs against existing methods.

Iterative Graph Seeking for Object Tracking.

To effectively solve the challenges in object tracking, such as large deformation and severe occlusion, many existing methods use graph-based models to capture target part relations, and adopt a sequential scheme of target part selection, part matching, and state estimation. However, such methods have two major drawbacks: 1) inaccurate part selection leads to performance deterioration of part matching and state estimation and 2) there are insufficient effective global constraints for local part selection and matching. In this paper, we propose a new object tracking method based on iterative graph seeking, which integrate target part selection, part matching, and state estimation using a unified energy minimization framework. Our method also incorporates structural information in local parts variations using the global constraint. We devise an alternative iteration scheme to minimize the energy function for searching the most plausible target geometric graph. Experimental results on several challenging benchmarks (i.e., VOT2015, OTB2013, and OTB2015) demonstrate improved performance and robustness in comparison with existing algorithms.

Geometric Hypergraph Learning for Visual Tracking.

Graph-based representation is widely used in visual tracking field by finding correct correspondences between target parts in different frames. However, most graph-based trackers consider pairwise geometric relations between local parts. They do not make full use of the target's intrinsic structure, thereby making the representation easily disturbed by errors in pairwise affinities when large deformation or occlusion occurs. In this paper, we propose a geometric hypergraph learning-based tracking method, which fully exploits high-order geometric relations among multiple correspondences of parts in different frames. Then visual tracking is formulated as the mode-seeking problem on the hypergraph in which vertices represent correspondence hypotheses and hyperedges describe high-order geometric relations among correspondences. Besides, a confidence-aware sampling method is developed to select representative vertices and hyperedges to construct the geometric hypergraph for more robustness and scalability. The experiments are carried out on three challenging datasets (VOT2014, OTB100, and Deform-SOT) to demonstrate that our method performs favorably against other existing trackers.

Online Deformable Object Tracking Based on Structure-Aware Hyper-Graph.

Recent advances in online visual tracking focus on designing part-based model to handle the deformation and occlusion challenges. However, previous methods usually consider only the pairwise structural dependences of target parts in two consecutive frames rather than the higher order constraints in multiple frames, making them less effective in handling large deformation and occlusion challenges. This paper describes a new and efficient method for online deformable object tracking. Different from most existing methods, this paper exploits higher order structural dependences of different parts of the tracking target in multiple consecutive frames. We construct a structure-aware hyper-graph to capture such higher order dependences, and solve the tracking problem by searching dense subgraphs on it. Furthermore, we also describe a new evaluating data set for online deformable object tracking (the Deform-SOT data set), which includes 50 challenging sequences with full annotations that represent realistic tracking challenges, such as large deformations and severe occlusions. The experimental result of the proposed method shows considerable improvement in performance over the state-of-the-art tracking methods.

An efficient reagent for covalent introduction of alkynes into proteins.

A cheap and bench-stable reagent was synthesized for direct and covalent introduction of alkynes into tyrosine of target proteins, which can be further modified based on click reaction to achieve fluorescence labelling or PEGylation. This reagent should be a generally useful toolbox for chemical biology and biomaterials.

Markov models for biogeography-based optimization.

Biogeography-based optimization (BBO) is a population-based evolutionary algorithm that is based on the mathematics of biogeography. Biogeography is the science and study of the geographical distribution of biological organisms. In BBO, problem solutions are analogous to islands, and the sharing of features between solutions is analogous to the migration of species. This paper derives Markov models for BBO with selection, migration, and mutation operators. Our models give the theoretically exact limiting probabilities for each possible population distribution for a given problem. We provide simulation results to confirm the Markov models.