arXiv subjects

Jie Li

Publications and source records attributed to Jie Li.

At least 595 records · Page 33Linked to original sources

arXiv2020

Collaborative Boundary-aware Context Encoding Networks for Error Map Prediction

Medical image segmentation is usually regarded as one of the most important intermediate steps in clinical situations and medical imaging research. Thus, accurately assessing the segmentation quality of the automatically generated predictions is essential for guaranteeing the reliability of the results of the computer-assisted diagnosis (CAD). Many researchers apply neural networks to train segmentation quality regression models to estimate the segmentation quality of a new data cohort without labeled ground truth. Recently, a novel idea is proposed that transforming the segmentation quality assessment (SQA) problem intothe pixel-wise error map prediction task in the form of segmentation. However, the simple application of vanilla segmentation structures in medical image fails to detect some small and thin error regions of the auto-generated masks with complex anatomical structures. In this paper, we propose collaborative boundaryaware context encoding networks called AEP-Net for error prediction task. Specifically, we propose a collaborative feature transformation branch for better feature fusion between images and masks, and precise localization of error regions. Further, we propose a context encoding module to utilize the global predictor from the error map to enhance the feature representation and regularize the networks. We perform experiments on IBSR v2.0 dataset and ACDC dataset. The AEP-Net achieves an average DSC of 0.8358, 0.8164 for error prediction task,and shows a high Pearson correlation coefficient of 0.9873 between the actual segmentation accuracy and the predicted accuracy inferred from the predicted error map on IBSR v2.0 dataset, which verifies the efficacy of our AEP-Net.

eess.IV↗

arXiv2020

Stationary quantum entanglement between a massive mechanical membrane and a low frequency LC circuit

We study electro-mechanical entanglement in a system where a massive membrane is capacitively coupled to a {\it low frequency} LC resonator. In opto- and electro-mechanics, the entanglement between a megahertz (MHz) mechanical resonator and a gigahertz (GHz) microwave LC resonator has been widely and well explored, and recently experimentally demonstrated. Typically, coupling is realized through a radiation pressure-like interaction, and entanglement is generated by adopting an appropriate microwave drive. Through this approach it is however not evident how to create entanglement in the case where both the mechanical and LC oscillators are of low frequency, e.g., around 1 MHz. Here we provide an effective approach to entangling two low-frequency resonators by further coupling the membrane to an optical cavity. The cavity is strongly driven by a red-detuned laser, sequentially cooling the mechanical and electrical modes, which results in stationary electro-mechanical entanglement at experimentally achievable temperatures. The entanglement directly originates from the electro-mechanical coupling itself and due to its quantum nature will allow testing quantum theories at a more macroscopic scale than currently possible.

quant-ph↗

arXiv2020

Multi-Margin based Decorrelation Learning for Heterogeneous Face Recognition

Heterogeneous face recognition (HFR) refers to matching face images acquired from different domains with wide applications in security scenarios. This paper presents a deep neural network approach namely Multi-Margin based Decorrelation Learning (MMDL) to extract decorrelation representations in a hyperspherical space for cross-domain face images. The proposed framework can be divided into two components: heterogeneous representation network and decorrelation representation learning. First, we employ a large scale of accessible visual face images to train heterogeneous representation network. The decorrelation layer projects the output of the first component into decorrelation latent subspace and obtains decorrelation representation. In addition, we design a multi-margin loss (MML), which consists of quadruplet margin loss (QML) and heterogeneous angular margin loss (HAML), to constrain the proposed framework. Experimental results on two challenging heterogeneous face databases show that our approach achieves superior performance on both verification and recognition tasks, comparing with state-of-the-art methods.

cs.CV↗

arXiv2020

Magnetostrictively induced stationary entanglement between two microwave fields

We present a scheme to entangle two microwave fields by using the nonlinear magnetostrictive interaction in a ferrimagnet. The magnetostrictive interaction enables the coupling between a magnon mode (spin wave) and a mechanical mode in the ferrimagnet, and the magnon mode simultaneously couples to two microwave cavity fields via the magnetic dipole interaction. The magnon-phonon coupling is enhanced by directly driving the ferrimagnet with a strong red-detuned microwave field, and the driving photons are scattered onto two sidebands induced by the mechanical motion. We show that two cavity fields can be prepared in a stationary entangled state if they are respectively resonant with two mechanical sidebands. The present scheme illustrates a new mechanism for creating entangled states of optical fields, and enables potential applications in quantum information science and quantum tasks that require entangled microwave fields.

quant-ph↗

arXiv2020

Projection & Probability-Driven Black-Box Attack

Generating adversarial examples in a black-box setting retains a significant challenge with vast practical application prospects. In particular, existing black-box attacks suffer from the need for excessive queries, as it is non-trivial to find an appropriate direction to optimize in the high-dimensional space. In this paper, we propose Projection & Probability-driven Black-box Attack (PPBA) to tackle this problem by reducing the solution space and providing better optimization. For reducing the solution space, we first model the adversarial perturbation optimization problem as a process of recovering frequency-sparse perturbations with compressed sensing, under the setting that random noise in the low-frequency space is more likely to be adversarial. We then propose a simple method to construct a low-frequency constrained sensing matrix, which works as a plug-and-play projection matrix to reduce the dimensionality. Such a sensing matrix is shown to be flexible enough to be integrated into existing methods like NES and Bandits$_{TD}$. For better optimization, we perform a random walk with a probability-driven strategy, which utilizes all queries over the whole progress to make full use of the sensing matrix for a less query budget. Extensive experiments show that our method requires at most 24% fewer queries with a higher attack success rate compared with state-of-the-art approaches. Finally, the attack method is evaluated on the real-world online service, i.e., Google Cloud Vision API, which further demonstrates our practical potentials.

cs.CV↗

arXiv2020

Anisotropic Convolutional Networks for 3D Semantic Scene Completion

As a voxel-wise labeling task, semantic scene completion (SSC) tries to simultaneously infer the occupancy and semantic labels for a scene from a single depth and/or RGB image. The key challenge for SSC is how to effectively take advantage of the 3D context to model various objects or stuffs with severe variations in shapes, layouts and visibility. To handle such variations, we propose a novel module called anisotropic convolution, which properties with flexibility and power impossible for the competing methods such as standard 3D convolution and some of its variations. In contrast to the standard 3D convolution that is limited to a fixed 3D receptive field, our module is capable of modeling the dimensional anisotropy voxel-wisely. The basic idea is to enable anisotropic 3D receptive field by decomposing a 3D convolution into three consecutive 1D convolutions, and the kernel size for each such 1D convolution is adaptively determined on the fly. By stacking multiple such anisotropic convolution modules, the voxel-wise modeling capability can be further enhanced while maintaining a controllable amount of model parameters. Extensive experiments on two SSC benchmarks, NYU-Depth-v2 and NYUCAD, show the superior performance of the proposed method. Our code is available at https://waterljwant.github.io/SSC/

cs.CV↗

arXiv2020

Real-Time Panoptic Segmentation from Dense Detections

Panoptic segmentation is a complex full scene parsing task requiring simultaneous instance and semantic segmentation at high resolution. Current state-of-the-art approaches cannot run in real-time, and simplifying these architectures to improve efficiency severely degrades their accuracy. In this paper, we propose a new single-shot panoptic segmentation network that leverages dense detections and a global self-attention mechanism to operate in real-time with performance approaching the state of the art. We introduce a novel parameter-free mask construction method that substantially reduces computational complexity by efficiently reusing information from the object detection and semantic segmentation sub-tasks. The resulting network has a simple data flow that does not require feature map re-sampling or clustering post-processing, enabling significant hardware acceleration. Our experiments on the Cityscapes and COCO benchmarks show that our network works at 30 FPS on 1024x2048 resolution, trading a 3% relative performance degradation from the current state of the art for up to 440% faster inference.

cs.CV↗

arXiv2020

Angular momentum-related probe of cold gas deficiencies

Recent studies of neutral atomic hydrogen (HI) in nearby galaxies found that all field disk galaxies are HI saturated, in that they carry roughly as much HI as permitted before this gas becomes gravitationally unstable. By taking this HI saturation for granted, the atomic gas fraction $f_{\rm atm}$ of galactic disks can be predicted as a function of the stability parameter $q=jσ/(GM)$, where $M$ and $j$ are the baryonic mass and specific angular momentum of the disk and $σ$ is the HI velocity dispersion Obreschkow et al. 2016. The log-ratio $Δf_q$ between this predictor and the observed atomic fraction can be seen as a physically motivated `HI deficiency'. While field disk galaxies have $Δf_q \approx0$, objects subject to environmental removal of HI are expected to have $Δf_q>0$. Within this framework, we revisit the HI deficiencies of satellite galaxies in the Virgo cluster and in clusters of the EAGLE simulation. We find that observed and simulated cluster galaxies are HI deficient and that $Δf_q$ slightly increases when getting closer to the cluster centres. The $Δf_q$ values are similar to traditional HI deficiency estimators, but $Δf_q$ is more directly comparable between observations and simulations than morphology-based deficiency estimators. By tracking the simulated HI deficient cluster galaxies back in time, we confirm that $Δf_q\approx0$ until the galaxies first enter a halo with $M_{\rm halo}>10^{13} {\rm M_{\odot}}$, at which moment they quickly lose HI by environmental effects. Finally, we use the simulation to investigate the links between $Δf_q$ and quenching of star formation.

astro-ph.GA↗

arXiv2020

Knowledge Mapping in Electricity Demand Forecasting: A Scientometric Insight

Forecasting electricity demand plays a fundamental role in the operation and planning procedures of power systems and the publications about electricity demand forecasting increasing year by year. In this paper, we use Scientometric analysis to analyze the current state and the emerging trends in the field of electricity demand forecasting form 831 publications of web of science core collection during 20 years: 1999-2018. Employing statistical description, cooperative network analysis, keyword co-occurrence analysis, co-citation analysis, cluster analysis, and emerging trend analysis techniques, this article gives the most critical countries, institutions, journals, authors and publications in this field, cooperative networks relationships, research hotspots and emerging trends. The results of this article can provide meaningful guidance and some insights for researchers to find out crucial research, emerging trends and new developments in this area.

physics.soc-ph↗

arXiv2020

Age of Information for Multicast Transmission with Fixed and Random Deadlines in IoT Systems

In this paper, we consider the multicast transmission of a real-time Internet of Things (IoT) system, where an access point (AP) transmits time-stamped status updates to multiple IoT devices. Different from the existing studies that only considered multicast transmission without deadlines, we enforce a deadline for the service time of each multicast status update, taking into account both the fixed and randomly distributed deadlines. In particular, a status update is dropped when either its deadline expires or it is successfully received by a certain number of IoT devices. Considering deadlines is important for many emerging IoT applications, where the outdated status updates are of no use to IoT devices. We evaluate the timeliness of the status update delivery by applying a recently proposed metric, named the age of information (AoI), which is defined as the time elapsed since the generation of the most recently received status update. After deriving the distributions of the service time for all possible reception outcomes at IoT devices, we manage to obtain the closed-form expressions of both the average AoI and the average peak AoI. Simulations validate the performance analysis, which reveals that the multicast transmission with deadlines achieves a lower average AoI than that without deadlines and there exists an optimal value of the deadline that can minimize the average (peak) AoI. Results also show that the fixed and random deadlines have respective advantages in different deadline regimes.

cs.IT↗

arXiv2020

Towards Practical Private Information Retrieval from MDS Array Codes

Private information retrieval (PIR) is the problem of privately retrieving one out of $M$ original files from $N$ severs, i.e., each individual server learns nothing about the file that the user is requesting. Usually, the $M$ files are replicated or encoded by a maximum distance separable (MDS) code and then stored across the $N$ servers. Compared to mere replication, MDS coded servers can significantly reduce the storage overhead. Particularly, PIR from minimum storage regenerating (MSR) coded servers can simultaneously reduce the repair bandwidth when repairing failed servers. Existing PIR schemes from MSR coded servers either require large sub-packetization levels or are not capacity-achieving. In this paper, a PIR protocol from MDS array codes is proposed, subsuming PIR from MSR coded servers as a special case. Particularly, the case of non-colluding, honest-but-curious servers is considered. The retrieval rate of the new PIR protocol achieves the capacity of PIR from MDS/MSR coded servers. By choosing different MDS array codes, the new PIR protocol can have some advantages when compared with existing protocols, e.g., 1) small sub-packetization, 2) (near-) optimal repair bandwidth, 3) implementable over the binary field $\mathbf{F}_2$.

cs.IT↗

arXiv2020

Learning to Make Chemical Predictions: the Interplay of Feature Representation, Data, and Machine Learning Algorithms

Recently supervised machine learning has been ascending in providing new predictive approaches for chemical, biological and materials sciences applications. In this Perspective we focus on the interplay of machine learning algorithm with the chemically motivated descriptors and the size and type of data sets needed for molecular property prediction. Using Nuclear Magnetic Resonance chemical shift prediction as an example, we demonstrate that success is predicated on the choice of feature extracted or real-space representations of chemical structures, whether the molecular property data is abundant and/or experimentally or computationally derived, and how these together will influence the correct choice of popular machine learning algorithms drawn from deep learning, random forests, or kernel methods.

physics.chem-ph↗

arXiv2020

Deep Learning for Feynman's Path Integral in Strong-Field Time-Dependent Dynamics

Feynman's path integral approach is to sum over all possible spatio-temporal paths to reproduce the quantum wave function and the corresponding time evolution, which has enormous potential to reveal quantum processes in classical view. However, the complete characterization of quantum wave function with infinite paths is a formidable challenge, which greatly limits the application potential, especially in the strong-field physics and attosecond science. Instead of brute-force tracking every path one by one, here we propose deep-learning-performed strong-field Feynman's formulation with pre-classification scheme which can predict directly the final results only with data of initial conditions, so as to attack unsurmountable tasks by existing strong-field methods and explore new physics. Our results build up a bridge between deep learning and strong-field physics through the Feynman's path integral, which would boost applications of deep learning to study the ultrafast time-dependent dynamics in strong-field physics and attosecond science, and shed a new light on the quantum-classical correspondence.

physics.atom-ph↗

arXiv2020

A Big Data Enabled Channel Model for 5G Wireless Communication Systems

The standardization process of the fifth generation (5G) wireless communications has recently been accelerated and the first commercial 5G services would be provided as early as in 2018. The increasing of enormous smartphones, new complex scenarios, large frequency bands, massive antenna elements, and dense small cells will generate big datasets and bring 5G communications to the era of big data. This paper investigates various applications of big data analytics, especially machine learning algorithms in wireless communications and channel modeling. We propose a big data and machine learning enabled wireless channel model framework. The proposed channel model is based on artificial neural networks (ANNs), including feed-forward neural network (FNN) and radial basis function neural network (RBF-NN). The input parameters are transmitter (Tx) and receiver (Rx) coordinates, Tx-Rx distance, and carrier frequency, while the output parameters are channel statistical properties, including the received power, root mean square (RMS) delay spread (DS), and RMS angle spreads (ASs). Datasets used to train and test the ANNs are collected from both real channel measurements and a geometry based stochastic model (GBSM). Simulation results show good performance and indicate that machine learning algorithms can be powerful analytical tools for future measurement-based wireless channel modeling.

eess.SP↗

arXiv2020

Semantically-Guided Representation Learning for Self-Supervised Monocular Depth

Self-supervised learning is showing great promise for monocular depth estimation, using geometry as the only source of supervision. Depth networks are indeed capable of learning representations that relate visual appearance to 3D properties by implicitly leveraging category-level patterns. In this work we investigate how to leverage more directly this semantic structure to guide geometric representation learning, while remaining in the self-supervised regime. Instead of using semantic labels and proxy losses in a multi-task approach, we propose a new architecture leveraging fixed pretrained semantic segmentation networks to guide self-supervised representation learning via pixel-adaptive convolutions. Furthermore, we propose a two-stage training process to overcome a common semantic bias on dynamic objects via resampling. Our method improves upon the state of the art for self-supervised monocular depth prediction over all pixels, fine-grained details, and per semantic categories.

cs.CV↗

arXiv2020

Joint Communication and Computational Resource Allocation for QoE-driven Point Cloud Video Streaming

Point cloud video is the most popular representation of hologram, which is the medium to precedent natural content in VR/AR/MR and is expected to be the next generation video. Point cloud video system provides users immersive viewing experience with six degrees of freedom and has wide applications in many fields such as online education, entertainment. To further enhance these applications, point cloud video streaming is in critical demand. The inherent challenges lie in the large size by the necessity of recording the three-dimensional coordinates besides color information, and the associated high computation complexity of encoding. To this end, this paper proposes a communication and computation resource allocation scheme for QoE-driven point cloud video streaming. In particular, we maximize system resource utilization by selecting different quantities, transmission forms and quality level tiles to maximize the quality of experience. Extensive simulations are conducted and the simulation results show the superior performance over the existing schemes

cs.MM↗

arXiv2020

3D Gated Recurrent Fusion for Semantic Scene Completion

This paper tackles the problem of data fusion in the semantic scene completion (SSC) task, which can simultaneously deal with semantic labeling and scene completion. RGB images contain texture details of the object(s) which are vital for semantic scene understanding. Meanwhile, depth images capture geometric clues of high relevance for shape completion. Using both RGB and depth images can further boost the accuracy of SSC over employing one modality in isolation. We propose a 3D gated recurrent fusion network (GRFNet), which learns to adaptively select and fuse the relevant information from depth and RGB by making use of the gate and memory modules. Based on the single-stage fusion, we further propose a multi-stage fusion strategy, which could model the correlations among different stages within the network. Extensive experiments on two benchmark datasets demonstrate the superior performance and the effectiveness of the proposed GRFNet for data fusion in SSC. Code will be made available.

cs.CV↗

arXiv2020

Facial Attribute Capsules for Noise Face Super Resolution

Existing face super-resolution (SR) methods mainly assume the input image to be noise-free. Their performance degrades drastically when applied to real-world scenarios where the input image is always contaminated by noise. In this paper, we propose a Facial Attribute Capsules Network (FACN) to deal with the problem of high-scale super-resolution of noisy face image. Capsule is a group of neurons whose activity vector models different properties of the same entity. Inspired by the concept of capsule, we propose an integrated representation model of facial information, which named Facial Attribute Capsule (FAC). In the SR processing, we first generated a group of FACs from the input LR face, and then reconstructed the HR face from this group of FACs. Aiming to effectively improve the robustness of FAC to noise, we generate FAC in semantic, probabilistic and facial attributes manners by means of integrated learning strategy. Each FAC can be divided into two sub-capsules: Semantic Capsule (SC) and Probabilistic Capsule (PC). Them describe an explicit facial attribute in detail from two aspects of semantic representation and probability distribution. The group of FACs model an image as a combination of facial attribute information in the semantic space and probabilistic space by an attribute-disentangling way. The diverse FACs could better combine the face prior information to generate the face images with fine-grained semantic attributes. Extensive benchmark experiments show that our method achieves superior hallucination results and outperforms state-of-the-art for very low resolution (LR) noise face image super resolution.

cs.CV↗