arXiv subjects

Jie Mei

Publications and source records attributed to Jie Mei.

65 records · Page 4Linked to original sources

arXiv2021

Absolute 3D Pose Estimation and Length Measurement of Severely Deformed Fish from Monocular Videos in Longline Fishing

Monocular absolute 3D fish pose estimation allows for efficient fish length measurement in the longline fisheries, where fishes are under severe deformation during the catching process. This task is challenging since it requires locating absolute 3D fish keypoints based on a short monocular video clip. Unlike related works, which either require expensive 3D ground-truth data and/or multiple-view images to provide depth information, or are limited to rigid objects, we propose a novel frame-based method to estimate the absolute 3D fish pose and fish length from a single-view 2D segmentation mask. We first introduce a relative 3D fish template. By minimizing an objective function, our method systematically estimates the relative 3D pose of the target fish and fish 2D keypoints in the image. Finally, with a closed-form solution, the relative 3D fish pose can help locate absolute 3D keypoints, resulting in the frame-based absolute fish length measurement, which is further refined based on the statistical temporal inference for the optimal fish length measurement from the video clip. Our experiments show that this method can accurately estimate the absolute 3D fish pose and further measure the absolute length, even outperforming the state-of-the-art multi-view method.

cs.CV↗

arXiv2021

Video-based Hierarchical Species Classification for Longline Fishing Monitoring

The goal of electronic monitoring (EM) of longline fishing is to monitor the fish catching activities on fishing vessels, either for the regulatory compliance or catch counting. Hierarchical classification based on videos allows for inexpensive and efficient fish species identification of catches from longline fishing, where fishes are under severe deformation and self-occlusion during the catching process. More importantly, the flexibility of hierarchical classification mitigates the laborious efforts of human reviews by providing confidence scores in different hierarchical levels. Some related works either use cascaded models for hierarchical classification or make predictions per image or predict one overlapping hierarchical data structure of the dataset in advance. However, with a known non-overlapping hierarchical data structure provided by fisheries scientists, our method enforces the hierarchical data structure and introduces an efficient training and inference strategy for video-based fisheries data. Our experiments show that the proposed method outperforms the classic flat classification system significantly and our ablation study justifies our contributions in CNN model design, training strategy, and the video-based inference schemes for the hierarchical fish species classification task.

cs.CV↗

arXiv2020

Inverse catalysis effect of quark anomalous magnetic moment to chiral restoration and deconfinement phase transitions

The effect of quark anomalous magnetic moment (AMM) to chiral restoration and deconfinement phase transitions under magnetic fields is investigated in a Pauli-Villars regularized PNJL model. A linear-in-$B$ term for quark anomalous magnetic moment is introduced to the Lagrangian density of our model, and it plays the role of inverse catalysis to the phase transitions. With fixed magnetic field, the critical temperature decreases with quark AMM. When fixing quark AMM, the critical temperature increases with magnetic field for a small quark AMM, but decreases with magnetic field for a large quark AMM. The critical temperature of chiral restoration and deconfinement phase transitions is determined by the two competing factors, the catalysis effect of magnetic field and inverse catalysis of quark anomalous magnetic moment.

hep-ph↗

arXiv2020

Machine learning for the diagnosis of Parkinson's disease: A systematic review

Diagnosis of Parkinson's disease (PD) is commonly based on medical observations and assessment of clinical signs, including the characterization of a variety of motor symptoms. However, traditional diagnostic approaches may suffer from subjectivity as they rely on the evaluation of movements that are sometimes subtle to human eyes and therefore difficult to classify, leading to possible misclassification. In the meantime, early non-motor symptoms of PD may be mild and can be caused by many other conditions. Therefore, these symptoms are often overlooked, making diagnosis of PD at an early stage challenging. To address these difficulties and to refine the diagnosis and assessment procedures of PD, machine learning methods have been implemented for the classification of PD and healthy controls or patients with similar clinical presentations (e.g., movement disorders or other Parkinsonian syndromes). To provide a comprehensive overview of data modalities and machine learning methods that have been used in the diagnosis and differential diagnosis of PD, in this study, we conducted a systematic literature review of studies published until February 14, 2020, using the PubMed and IEEE Xplore databases. A total of 209 studies were included, extracted for relevant information and presented in this systematic review, with an investigation of their aims, sources of data, types of data, machine learning methods and associated outcomes. These studies demonstrate a high potential for adaptation of machine learning methods and novel biomarkers in clinical decision making, leading to increasingly systematic, informed diagnosis of PD.

cs.LG↗

arXiv2020

SLCRF: Subspace Learning with Conditional Random Field for Hyperspectral Image Classification

Subspace learning (SL) plays an important role in hyperspectral image (HSI) classification, since it can provide an effective solution to reduce the redundant information in the image pixels of HSIs. Previous works about SL aim to improve the accuracy of HSI recognition. Using a large number of labeled samples, related methods can train the parameters of the proposed solutions to obtain better representations of HSI pixels. However, the data instances may not be sufficient enough to learn a precise model for HSI classification in real applications. Moreover, it is well-known that it takes much time, labor and human expertise to label HSI images. To avoid the aforementioned problems, a novel SL method that includes the probability assumption called subspace learning with conditional random field (SLCRF) is developed. In SLCRF, first, the 3D convolutional autoencoder (3DCAE) is introduced to remove the redundant information in HSI pixels. In addition, the relationships are also constructed using the spectral-spatial information among the adjacent pixels. Then, the conditional random field (CRF) framework can be constructed and further embedded into the HSI SL procedure with the semi-supervised approach. Through the linearized alternating direction method termed LADMAP, the objective function of SLCRF is optimized using a defined iterative algorithm. The proposed method is comprehensively evaluated using the challenging public HSI datasets. We can achieve stateof-the-art performance using these HSI sets.

cs.CV↗

arXiv2019

Intelligent Network Slicing for V2X Services Towards 5G

Benefiting from the widely deployed LTE infrastructures, the fifth generation (5G) wireless networks have been becoming a critical enabler for the emerging vehicle-to-everything (V2X) communications. However, existing LTE networks cannot efficiently support stringent but dynamic requirements of V2X services. One effective solution to overcome this challenge is network slicing, whereby different services could be supported by logically separated networks. To mitigate the increasing complexity of network slicing in 5G, we propose to leverage the recent advancement of Machine Learning (ML) technologies for automated network operation. Specifically, we propose intelligent network slicing architecture for V2X services, where network functions and multi-dimensional network resources are virtualized and assigned to different network slices. In achieving optimized slicing intelligently, several critical techniques, including mobile data collection and ML algorithm design, are discussed to tackle the related challenges. Then, we develop a simulation platform to illustrate the effectiveness of our proposed intelligent network slicing. With integration of 5G network slicing and ML-enabled technologies, the QoS of V2X services is expected to be dramatically enhanced.

cs.NI↗

arXiv2019

Distributed Consensus for Multiple Lagrangian Systems with Parametric Uncertainties and External Disturbances Under Directed Graphs

In this paper, we study the leaderless consensus problem for multiple Lagrangian systems in the presence of parametric uncertainties and external disturbances under directed graphs. For achieving asymptotic behavior, a robust continuous term with adaptive varying gains is added to alleviate the effects of the external disturbances with unknown bounds. In the case of a fixed directed graph, by introducing an integrate term in the auxiliary variable design, the final consensus equilibrium can be explicitly derived. We show that the agents achieve weighted average consensus, where the final equilibrium is dependent on three factors, namely, the interactive topology, the initial positions of the agents, and the control gains of the proposed control algorithm. In the case of switching directed graphs, a model reference adaptive consensus based algorithm is proposed such that the agents achieve leaderless consensus if the infinite sequence of switching graphs is uniformly jointly connected. Motivated by the fact that the relative velocity information is difficult to obtain accurately, we further propose a leaderless consensus algorithm with gain adaptation for multiple Lagrangian systems without using neighbors' velocity information. We also propose a model reference adaptive consensus based algorithm without using neighbors' velocity information for switching directed graphs. The proposed algorithms are distributed in the sense of using local information from its neighbors and using no comment control gains. Numerical simulations are performed to show the effectiveness of the proposed algorithms.

eess.SY↗

arXiv2019

Ultra-Reliable and Low-Latency Communications for Connected Vehicles: Challenges and Solutions

With the rapid development of communications and computing, the concept of connected vehicles emerges to improve driving safety, traffic efficiency and infotainment experience. Due to the limited capabilities of sensors and information processing on a single vehicle, vehicular networks (VNETs) play a vital role for the realization of connected vehicles. To achieve the closer cooperation on the road, ultra-reliable and low-latency communications (URLLCs) become the indispensable components of future VNETs. However, traditional wireless networks are designed with the objective of maximizing spectral and energy efficiency regardless of adequate consideration of URLLCs. To this end, this article investigates the challenges and solutions of conceiving URLLC VNETs dedicated for connected vehicles. Specifically, the underlying application scenarios are first presented with various vehicular communication patterns. Then, the URLLC requirements and potential challenges are discussed for VNETs. We also discuss the solutions to establish URLLC VNETs from the aspects of physical and media access control layers, including the optimization frameworks for latency and reliability, as well as the case studies of employing such frameworks. Finally, several open research directions are pointed out.

cs.IT↗

arXiv2018

Systems of Global Governance in the Era of Human-Machine Convergence

Technology is increasingly shaping our social structures and is becoming a driving force in altering human biology. Besides, human activities already proved to have a significant impact on the Earth system which in turn generates complex feedback loops between social and ecological systems. Furthermore, since our species evolved relatively fast from small groups of hunter-gatherers to large and technology-intensive urban agglomerations, it is not a surprise that the major institutions of human society are no longer fit to cope with the present complexity. In this note we draw foundational parallelisms between neurophysiological systems and ICT-enabled social systems, discussing how frameworks rooted in biology and physics could provide heuristic value in the design of evolutionary systems relevant to politics and economics. In this regard we highlight how the governance of emerging technology (i.e. nanotechnology, biotechnology, information technology, and cognitive science), and the one of climate change both presently confront us with a number of connected challenges. In particular: historically high level of inequality; the co-existence of growing multipolar cultural systems in an unprecedentedly connected world; the unlikely reaching of the institutional agreements required to deviate abnormal trajectories of development. We argue that wise general solutions to such interrelated issues should embed the deep understanding of how to elicit mutual incentives in the socio-economic subsystems of Earth system in order to jointly concur to a global utility function (e.g. avoiding the reach of planetary boundaries and widespread social unrest). We leave some open questions on how techno-social systems can effectively learn and adapt with respect to our understanding of geopolitical complexity.

cs.CY↗

arXiv2016

Statistical Learning for OCR Text Correction

The accuracy of Optical Character Recognition (OCR) is crucial to the success of subsequent applications used in text analyzing pipeline. Recent models of OCR post-processing significantly improve the quality of OCR-generated text, but are still prone to suggest correction candidates from limited observations while insufficiently accounting for the characteristics of OCR errors. In this paper, we show how to enlarge candidate suggestion space by using external corpus and integrating OCR-specific features in a regression approach to correct OCR-generated errors. The evaluation results show that our model can correct 61.5% of the OCR-errors (considering the top 1 suggestion) and 71.5% of the OCR-errors (considering the top 3 suggestions), for cases where the theoretical correction upper-bound is 78%.

cs.CV↗

arXiv2015

Fully Distributed Flocking with a Moving Leader for Lagrange Networks with Parametric Uncertainties

This paper addresses the leader-follower flocking problem with a moving leader for networked Lagrange systems with parametric uncertainties under a proximity graph. Here a group of followers move cohesively with the moving leader to maintain connectivity and avoid collisions for all time and also eventually achieve velocity matching. In the proximity graph, the neighbor relationship is defined according to the relative distance between each pair of agents. Each follower is able to obtain information from only the neighbors in its proximity, involving only local interaction. We consider two cases: i) the leader moves with a constant velocity, and ii) the leader moves with a varying velocity. In the first case, a distributed continuous adaptive control algorithm accounting for unknown parameters is proposed in combination with a distributed continuous estimator for each follower. In the second case, a distributed discontinuous adaptive control algorithm and estimator are proposed. Then the algorithm is extended to be fully distributed with the introduction of gain adaptation laws. In all proposed algorithms, only one-hop neighbors' information (e.g., the relative position and velocity measurements between the neighbors and the absolute position and velocity measurements) is required, and flocking is achieved as long as the connectivity and collision avoidance are ensured at the initial time and the control gains are designed properly. Numerical simulations are presented to illustrate the theoretical results.

math.OC↗