Searcharxiv⌕ Search

arXiv subjects

Jian Sun

Publications and source records attributed to Jian Sun.

At least 415 records · Page 23Linked to original sources
arXiv2019

Prediction of pressure-induced stabilization of noble-gas-atom compounds with alkali oxides and alkali sulfides

The cubic antifluorite structure comprises a face-centered cubic sublattice of anions with cations on the tetrahedral sites. The voids in the antifluorite structure that are crucial for superionicity in Li2O might also act as atomic traps. Trapping of guest atoms and small molecules within voids of a host structure leads to the formation of what are known as clathrate compounds. Here we investigate the possibility of trapping helium or larger neon guest atoms under pressure within alkali metal oxide and sulfide structures. We find stable helium and neon-bearing compounds at very low pressures. These structures are stabilized by a reduction in volume from incorporation of helium or neon atoms within the antifluorite structure. We predict that NeCs2S could be stable at ambient pressure. Our study suggests a novel class of alkali oxide and sulfide materials incorporating noble gas atoms that might potentially be useful for gas storage.

cond-mat.mtrl-sci↗
arXiv2018

Spatial multiplexing of squeezed light by coherence diffusion

Spatially splitting nonclassical light beams is in principle prohibited due to noise contamination during beam splitting. We propose a platform based on thermal motion of atoms to realize spatial multiplexing of squeezed light. Light channels of separate spatial modes in an anti-relaxation coated vapor cell share the same long-lived atomic coherence jointly created by all channels through the coherent diffusion of atoms which in turn enhances individual channel's nonlinear process responsible for light squeezing. Consequently, it behaves as squeezed light in one optical channel transferring to other distant channels even with laser powers below the threshold for squeezed light generation. An array of squeezed light beams was created with low laser power $\sim\text{mW}$. This approach holds great promise for applications in multi-node quantum network and quantum enhanced technologies such as quantum imaging and sensing.

quant-ph↗
arXiv2018

Sub-Hertz resonance by weak measurement

Weak measurement (WM) with state pre- and post-selection can amplify otherwise undetectable small signals and thus promise great potentials in precision measurements. Although frequency measurements offer the hitherto highest precision owing to stable narrow atomic transitions, it remains a long-standing interest to develop new schemes to further escalate their performance. Here, we propose and demonstrate a WM-enhanced spectroscopy technique which is capable of narrowing the resonance to 0.1 Hz in a room-temperature atomic vapor cell. Potential of this technique for precision measurement is demonstrated through weak magnetic field sensing. By judiciously pre- and post-selecting frequency-modulated input and output optical states in a nearly-orthogonal manner, a sensitivity of $\text{7 fT/}\sqrt{\text{Hz}}$ near DC is achieved, using only one laser beam of $\text{7 }\text{μW}$ power. Additionally, our results extend the WM framework to a non-Hermitian Hamiltonian, and shed new light in metrology and bio-magnetic field sensing applications.

physics.atom-ph↗
arXiv2018

HyperAdam: A Learnable Task-Adaptive Adam for Network Training

Deep neural networks are traditionally trained using human-designed stochastic optimization algorithms, such as SGD and Adam. Recently, the approach of learning to optimize network parameters has emerged as a promising research topic. However, these learned black-box optimizers sometimes do not fully utilize the experience in human-designed optimizers, therefore have limitation in generalization ability. In this paper, a new optimizer, dubbed as \textit{HyperAdam}, is proposed that combines the idea of "learning to optimize" and traditional Adam optimizer. Given a network for training, its parameter update in each iteration generated by HyperAdam is an adaptive combination of multiple updates generated by Adam with varying decay rates. The combination weights and decay rates in HyperAdam are adaptively learned depending on the task. HyperAdam is modeled as a recurrent neural network with AdamCell, WeightCell and StateCell. It is justified to be state-of-the-art for various network training, such as multilayer perceptron, CNN and LSTM.

cs.LG↗
arXiv2018

MetaAnchor: Learning to Detect Objects with Customized Anchors

We propose a novel and flexible anchor mechanism named MetaAnchor for object detection frameworks. Unlike many previous detectors model anchors via a predefined manner, in MetaAnchor anchor functions could be dynamically generated from the arbitrary customized prior boxes. Taking advantage of weight prediction, MetaAnchor is able to work with most of the anchor-based object detection systems such as RetinaNet. Compared with the predefined anchor scheme, we empirically find that MetaAnchor is more robust to anchor settings and bounding box distributions; in addition, it also shows the potential on transfer tasks. Our experiment on COCO detection task shows that MetaAnchor consistently outperforms the counterparts in various scenarios.

cs.CV↗
arXiv2018

Learning Spectral Transform Network on 3D Surface for Non-rigid Shape Analysis

Designing a network on 3D surface for non-rigid shape analysis is a challenging task. In this work, we propose a novel spectral transform network on 3D surface to learn shape descriptors. The proposed network architecture consists of four stages: raw descriptor extraction, surface second-order pooling, mixture of power function-based spectral transform, and metric learning. The proposed network is simple and shallow. Quantitative experiments on challenging benchmarks show its effectiveness for non-rigid shape retrieval and classification, e.g., it achieved the highest accuracies on SHREC14, 15 datasets as well as the Range subset of SHREC17 dataset.

cs.CV↗
arXiv2018

Unpaired Brain MR-to-CT Synthesis using a Structure-Constrained CycleGAN

The cycleGAN is becoming an influential method in medical image synthesis. However, due to a lack of direct constraints between input and synthetic images, the cycleGAN cannot guarantee structural consistency between these two images, and such consistency is of extreme importance in medical imaging. To overcome this, we propose a structure-constrained cycleGAN for brain MR-to-CT synthesis using unpaired data that defines an extra structure-consistency loss based on the modality independent neighborhood descriptor to constrain structural consistency. Additionally, we use a position-based selection strategy for selecting training images instead of a completely random selection scheme. Experimental results on synthesizing CT images from brain MR images demonstrate that our method is better than the conventional cycleGAN and approximates the cycleGAN trained with paired data.

cs.CV↗
arXiv2018

GridFace: Face Rectification via Learning Local Homography Transformations

In this paper, we propose a method, called GridFace, to reduce facial geometric variations and improve the recognition performance. Our method rectifies the face by local homography transformations, which are estimated by a face rectification network. To encourage the image generation with canonical views, we apply a regularization based on the natural face distribution. We learn the rectification network and recognition network in an end-to-end manner. Extensive experiments show our method greatly reduces geometric variations, and gains significant improvements in unconstrained face recognition scenarios.

cs.CV↗
arXiv2018

Neural Multi-Atlas Label Fusion: Application to Cardiac MR Images

Multi-atlas segmentation approach is one of the most widely-used image segmentation techniques in biomedical applications. There are two major challenges in this category of methods, i.e., atlas selection and label fusion. In this paper, we propose a novel multi-atlas segmentation method that formulates multi-atlas segmentation in a deep learning framework for better solving these challenges. The proposed method, dubbed deep fusion net (DFN), is a deep architecture that integrates a feature extraction subnet and a non-local patch-based label fusion (NL-PLF) subnet in a single network. The network parameters are learned by end-to-end training for automatically learning deep features that enable optimal performance in a NL-PLF framework. The learned deep features are further utilized in defining a similarity measure for atlas selection. By evaluating on two public cardiac MR datasets of SATA-13 and LV-09 for left ventricle segmentation, our approach achieved 0.833 in averaged Dice metric (ADM) on SATA-13 dataset and 0.95 in ADM for epicardium segmentation on LV-09 dataset, comparing favorably with the other automatic left ventricle segmentation methods. We also tested our approach on Cardiac Atlas Project (CAP) testing set of MICCAI 2013 SATA Segmentation Challenge, and our method achieved 0.815 in ADM, ranking highest at the time of writing.

cs.CV↗
arXiv2018

ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design

Currently, the neural network architecture design is mostly guided by the \emph{indirect} metric of computation complexity, i.e., FLOPs. However, the \emph{direct} metric, e.g., speed, also depends on the other factors such as memory access cost and platform characterics. Thus, this work proposes to evaluate the direct metric on the target platform, beyond only considering FLOPs. Based on a series of controlled experiments, this work derives several practical \emph{guidelines} for efficient network design. Accordingly, a new architecture is presented, called \emph{ShuffleNet V2}. Comprehensive ablation experiments verify that our model is the state-of-the-art in terms of speed and accuracy tradeoff.

cs.CV↗
arXiv2018

Unified Perceptual Parsing for Scene Understanding

Humans recognize the visual world at multiple levels: we effortlessly categorize scenes and detect objects inside, while also identifying the textures and surfaces of the objects along with their different compositional parts. In this paper, we study a new task called Unified Perceptual Parsing, which requires the machine vision systems to recognize as many visual concepts as possible from a given image. A multi-task framework called UPerNet and a training strategy are developed to learn from heterogeneous image annotations. We benchmark our framework on Unified Perceptual Parsing and show that it is able to effectively segment a wide range of concepts from images. The trained networks are further applied to discover visual knowledge in natural scenes. Models are available at \url{https://github.com/CSAILVision/unifiedparsing}.

cs.CV↗
arXiv2018

Learning Visually-Grounded Semantics from Contrastive Adversarial Samples

We study the problem of grounding distributional representations of texts on the visual domain, namely visual-semantic embeddings (VSE for short). Begin with an insightful adversarial attack on VSE embeddings, we show the limitation of current frameworks and image-text datasets (e.g., MS-COCO) both quantitatively and qualitatively. The large gap between the number of possible constitutions of real-world semantics and the size of parallel data, to a large extent, restricts the model to establish the link between textual semantics and visual concepts. We alleviate this problem by augmenting the MS-COCO image captioning datasets with textual contrastive adversarial samples. These samples are synthesized using linguistic rules and the WordNet knowledge base. The construction procedure is both syntax- and semantics-aware. The samples enforce the model to ground learned embeddings to concrete concepts within the image. This simple but powerful technique brings a noticeable improvement over the baselines on a diverse set of downstream tasks, in addition to defending known-type adversarial attacks. We release the codes at https://github.com/ExplorerFreda/VSE-C.

cs.CL↗
arXiv2018

Unusual phonon density of states and response to the superconducting transition in the In-doped topological crystalline insulator Pb$_{0.5}$Sn$_{0.5}$Te

We present inelastic neutron scattering results of phonons in (Pb$_{0.5}$Sn$_{0.5}$)$_{1-x}$In$_x$Te powders, with $x=0$ and 0.3. The $x=0$ sample is a topological crystalline insulator, and the $x=0.3$ sample is a superconductor with a bulk superconducting transition temperature $T_c$ of 4.7 K. In both samples, we observe unexpected van Hove singularities in the phonon density of states at energies of 1--2.5 meV, suggestive of local modes. On cooling the superconducting sample through $T_c$, there is an enhancement of these features for energies below twice the superconducting-gap energy. We further note that the superconductivity in (Pb$_{0.5}$Sn$_{0.5}$)$_{1-x}$In$_x$Te occurs in samples with normal-state resistivities of order 10 m$Ω$~cm, indicative of bad-metal behavior. Calculations based on density functional theory suggest that the superconductivity is easily explainable in terms of electron-phonon coupling; however, they completely miss the low-frequency modes and do not explain the large resistivity. While the bulk superconducting state of (Pb$_{0.5}$Sn$_{0.5}$)$_{0.7}$In$_{0.3}$Te appears to be driven by phonons, a proper understanding will require ideas beyond simple BCS theory.

cond-mat.supr-con↗
arXiv2018

Evidence for a Dirac nodal-line semimetal in SrAs$_{3}$

Dirac nodal-line semimetals with the linear bands crossing along a line or loop, represent a new topological state of matter. Here, by carrying out magnetotransport measurements and performing first-principle calculations, we demonstrate that such a state has been realized in high-quality single crystals of SrAs3. We obtain the nontrivial pi Berry phase by analysing the Shubnikov-de Haas quantum oscillations. We also observe a robust negative longitudinal magnetoresistance induced by the chiral anomaly. Accompanying first-principles calculations identify that a single hole pocket enclosing the loop nodes is responsible for these observations.

cond-mat.mtrl-sci↗
arXiv2018

CrowdHuman: A Benchmark for Detecting Human in a Crowd

Human detection has witnessed impressive progress in recent years. However, the occlusion issue of detecting human in highly crowded environments is far from solved. To make matters worse, crowd scenarios are still under-represented in current human detection benchmarks. In this paper, we introduce a new dataset, called CrowdHuman, to better evaluate detectors in crowd scenarios. The CrowdHuman dataset is large, rich-annotated and contains high diversity. There are a total of $470K$ human instances from the train and validation subsets, and $~22.6$ persons per image, with various kinds of occlusions in the dataset. Each human instance is annotated with a head bounding-box, human visible-region bounding-box and human full-body bounding-box. Baseline performance of state-of-the-art detection frameworks on CrowdHuman is presented. The cross-dataset generalization results of CrowdHuman dataset demonstrate state-of-the-art performance on previous dataset including Caltech-USA, CityPersons, and Brainwash without bells and whistles. We hope our dataset will serve as a solid baseline and help promote future research in human detection tasks.

cs.CV↗
arXiv2018

DetNet: A Backbone network for Object Detection

Recent CNN based object detectors, no matter one-stage methods like YOLO, SSD, and RetinaNe or two-stage detectors like Faster R-CNN, R-FCN and FPN are usually trying to directly finetune from ImageNet pre-trained models designed for image classification. There has been little work discussing on the backbone feature extractor specifically designed for the object detection. More importantly, there are several differences between the tasks of image classification and object detection. 1. Recent object detectors like FPN and RetinaNet usually involve extra stages against the task of image classification to handle the objects with various scales. 2. Object detection not only needs to recognize the category of the object instances but also spatially locate the position. Large downsampling factor brings large valid receptive field, which is good for image classification but compromises the object location ability. Due to the gap between the image classification and object detection, we propose DetNet in this paper, which is a novel backbone network specifically designed for object detection. Moreover, DetNet includes the extra stages against traditional backbone network for image classification, while maintains high spatial resolution in deeper layers. Without any bells and whistles, state-of-the-art results have been obtained for both object detection and instance segmentation on the MSCOCO benchmark based on our DetNet~(4.8G FLOPs) backbone. The code will be released for the reproduction.

cs.CV↗
arXiv2018

Automated vehicle's behavior decision making using deep reinforcement learning and high-fidelity simulation environment

Automated vehicles are deemed to be the key element for the intelligent transportation system in the future. Many studies have been made to improve the Automated vehicles' ability of environment recognition and vehicle control, while the attention paid to decision making is not enough though the decision algorithms so far are very preliminary. Therefore, a framework of the decision-making training and learning is put forward in this paper. It consists of two parts: the deep reinforcement learning training program and the high-fidelity virtual simulation environment. Then the basic microscopic behavior, car-following, is trained within this framework. In addition, theoretical analysis and experiments were conducted on setting reward function for accelerating training using deep reinforcement learning. The results show that on the premise of driving comfort, the efficiency of the trained Automated vehicle increases 7.9% compared to the classical traffic model, intelligent driver model. Later on, on a more complex three-lane section, we trained the integrated model combines both car-following and lane-changing behavior, the average speed further grows 2.4%. It indicates that our framework is effective for Automated vehicle's decision-making learning.

cs.AI↗