arXiv subjects

Zhen Sun

Publications and source records attributed to Zhen Sun.

67 records · Page 4Linked to original sources

arXiv2024

SMART: Scene-motion-aware human action recognition framework for mental disorder group

Patients with mental disorders often exhibit risky abnormal actions, such as climbing walls or hitting windows, necessitating intelligent video behavior monitoring for smart healthcare with the rising Internet of Things (IoT) technology. However, the development of vision-based Human Action Recognition (HAR) for these actions is hindered by the lack of specialized algorithms and datasets. In this paper, we innovatively propose to build a vision-based HAR dataset including abnormal actions often occurring in the mental disorder group and then introduce a novel Scene-Motion-aware Action Recognition Technology framework, named SMART, consisting of two technical modules. First, we propose a scene perception module to extract human motion trajectory and human-scene interaction features, which introduces additional scene information for a supplementary semantic representation of the above actions. Second, the multi-stage fusion module fuses the skeleton motion, motion trajectory, and human-scene interaction features, enhancing the semantic association between the skeleton motion and the above supplementary representation, thus generating a comprehensive representation with both human motion and scene information. The effectiveness of our proposed method has been validated on our self-collected HAR dataset (MentalHAD), achieving 94.9% and 93.1% accuracy in un-seen subjects and scenes and outperforming state-of-the-art approaches by 6.5% and 13.2%, respectively. The demonstrated subject- and scene- generalizability makes it possible for SMART's migration to practical deployment in smart healthcare systems for mental disorder patients in medical settings. The code and dataset will be released publicly for further research: https://github.com/Inowlzy/SMART.git.

cs.CV↗

arXiv2024

TON-VIO: Online Time Offset Modeling Networks for Robust Temporal Alignment in High Dynamic Motion VIO

Temporal misalignment (time offset) between sensors is common in low cost visual-inertial odometry (VIO) systems. Such temporal misalignment introduces inconsistent constraints for state estimation, leading to a significant positioning drift especially in high dynamic motion scenarios. In this article, we focus on online temporal calibration to reduce the positioning drift caused by the time offset for high dynamic motion VIO. For the time offset observation model, most existing methods rely on accurate state estimation or stable visual tracking. For the prediction model, current methods oversimplify the time offset as a constant value with white Gaussian noise. However, these ideal conditions are seldom satisfied in real high dynamic scenarios, resulting in the poor performance. In this paper, we introduce online time offset modeling networks (TON) to enhance real-time temporal calibration. TON improves the accuracy of time offset observation and prediction modeling. Specifically, for observation modeling, we propose feature velocity observation networks to enhance velocity computation for features in unstable visual tracking conditions. For prediction modeling, we present time offset prediction networks to learn its evolution pattern. To highlight the effectiveness of our method, we integrate the proposed TON into both optimization-based and filter-based VIO systems. Simulation and real-world experiments are conducted to demonstrate the enhanced performance of our approach. Additionally, to contribute to the VIO community, we will open-source the code of our method on: https://github.com/Franky-X/FVON-TPN.

cs.RO↗

arXiv2023

PWISeg: Point-based Weakly-supervised Instance Segmentation for Surgical Instruments

In surgical procedures, correct instrument counting is essential. Instance segmentation is a location method that locates not only an object's bounding box but also each pixel's specific details. However, obtaining mask-level annotations is labor-intensive in instance segmentation. To address this issue, we propose a novel yet effective weakly-supervised surgical instrument instance segmentation approach, named Point-based Weakly-supervised Instance Segmentation (PWISeg). PWISeg adopts an FCN-based architecture with point-to-box and point-to-mask branches to model the relationships between feature points and bounding boxes, as well as feature points and segmentation masks on FPN, accomplishing instrument detection and segmentation jointly in a single model. Since mask level annotations are hard to available in the real world, for point-to-mask training, we introduce an unsupervised projection loss, utilizing the projected relation between predicted masks and bboxes as supervision signal. On the other hand, we annotate a few pixels as the key pixel for each instrument. Based on this, we further propose a key pixel association loss and a key pixel distribution loss, driving the point-to-mask branch to generate more accurate segmentation predictions. To comprehensively evaluate this task, we unveil a novel surgical instrument dataset with manual annotations, setting up a benchmark for further research. Our comprehensive research trial validated the superior performance of our PWISeg. The results show that the accuracy of surgical instrument segmentation is improved, surpassing most methods of instance segmentation via weakly supervised bounding boxes. This improvement is consistently observed in our proposed dataset and when applied to the public HOSPI-Tools dataset.

cs.CV↗

arXiv2023

Janus icosahedral particles: amorphization driven by three-dimensional atomic misfit and edge dislocation compensation

Icosahedral nanoparticles composed of fivefold twinned tetrahedra have broad applications. The strain relief mechanism and angular deficiency in icosahedral multiply twinned particles are poorly understood in three dimensions. Here, we resolved the three-dimensional atomic structures of Janus icosahedral nanoparticles using atomic resolution electron tomography. A geometrically fivefold face consistently corresponds to a less ordered face like two hemispheres. We quantify rich structural variety of icosahedra including bond orientation order, bond length, strain tensor; and packing efficiency, atom number, solid angle of each tetrahedron. These structural characteristics exhibit two-sided distribution. Edge dislocations near the axial atoms and small disordered domains fill the angular deficiency. Our findings provide new insights how the fivefold symmetry can be compensated and the geometrically-necessary internal strains relived in multiply twinned particles.

cond-mat.mtrl-sci↗

arXiv2023

NeRF-LOAM: Neural Implicit Representation for Large-Scale Incremental LiDAR Odometry and Mapping

Simultaneously odometry and mapping using LiDAR data is an important task for mobile systems to achieve full autonomy in large-scale environments. However, most existing LiDAR-based methods prioritize tracking quality over reconstruction quality. Although the recently developed neural radiance fields (NeRF) have shown promising advances in implicit reconstruction for indoor environments, the problem of simultaneous odometry and mapping for large-scale scenarios using incremental LiDAR data remains unexplored. To bridge this gap, in this paper, we propose a novel NeRF-based LiDAR odometry and mapping approach, NeRF-LOAM, consisting of three modules neural odometry, neural mapping, and mesh reconstruction. All these modules utilize our proposed neural signed distance function, which separates LiDAR points into ground and non-ground points to reduce Z-axis drift, optimizes odometry and voxel embeddings concurrently, and in the end generates dense smooth mesh maps of the environment. Moreover, this joint optimization allows our NeRF-LOAM to be pre-trained free and exhibit strong generalization abilities when applied to different environments. Extensive evaluations on three publicly available datasets demonstrate that our approach achieves state-of-the-art odometry and mapping performance, as well as a strong generalization in large-scale environments utilizing LiDAR data. Furthermore, we perform multiple ablation studies to validate the effectiveness of our network design. The implementation of our approach will be made available at https://github.com/JunyuanDeng/NeRF-LOAM.

cs.CV↗

arXiv2021

Generic Security Analysis Framework for Quantum Secure Direct Communication

Quantum secure direct communication provides a direct means of conveying secret information via quantum states among legitimate users. The past two decades have witnessed its great strides both theoretically and experimentally. However, the security analysis of it still stays in its infant. Some practical problems in this field to be solved urgently, such as detector efficiency mismatch, side-channel effect and source imperfection, are propelling the birth of a more impeccable solution. In this paper, we establish a new framework of the security analysis driven by numerics where all the practical problems may be taken into account naturally. We apply this framework to several variations of the DL04 protocol considering real-world experimental conditions. Also, we propose two optimizing methods to process the numerical part of the framework so as to meet different requirements in practice. With these properties considered, we predict the robust framework would open up a broad avenue of the development in the field.

quant-ph↗

arXiv2020

De-Anonymizing Text by Fingerprinting Language Generation

Components of machine learning systems are not (yet) perceived as security hotspots. Secure coding practices, such as ensuring that no execution paths depend on confidential inputs, have not yet been adopted by ML developers. We initiate the study of code security of ML systems by investigating how nucleus sampling---a popular approach for generating text, used for applications such as auto-completion---unwittingly leaks texts typed by users. Our main result is that the series of nucleus sizes for many natural English word sequences is a unique fingerprint. We then show how an attacker can infer typed text by measuring these fingerprints via a suitable side channel (e.g., cache access times), explain how this attack could help de-anonymize anonymous texts, and discuss defenses.

cs.CR↗

arXiv2020

Attention-SLAM: A Visual Monocular SLAM Learning from Human Gaze

This paper proposes a novel simultaneous localization and mapping (SLAM) approach, namely Attention-SLAM, which simulates human navigation mode by combining a visual saliency model (SalNavNet) with traditional monocular visual SLAM. Most SLAM methods treat all the features extracted from the images as equal importance during the optimization process. However, the salient feature points in scenes have more significant influence during the human navigation process. Therefore, we first propose a visual saliency model called SalVavNet in which we introduce a correlation module and propose an adaptive Exponential Moving Average (EMA) module. These modules mitigate the center bias to enable the saliency maps generated by SalNavNet to pay more attention to the same salient object. Moreover, the saliency maps simulate the human behavior for the refinement of SLAM results. The feature points extracted from the salient regions have greater importance in optimization process. We add semantic saliency information to the Euroc dataset to generate an open-source saliency SLAM dataset. Comprehensive test results prove that Attention-SLAM outperforms benchmarks such as Direct Sparse Odometry (DSO), ORB-SLAM, and Salient DSO in terms of efficiency, accuracy, and robustness in most test cases.

cs.CV↗

arXiv2020

Experimental free-space quantum secure direct communication and its security analysis

We report an experimental implementation of free-space quantum secure direct communication based on single photons. The quantum communication scheme uses phase encoding, and the asymmetric Mach-Zehnder interferometer is optimized so as to automatically compensate phase drift of the photons during their transitions over the free-space medium. An information transmission rate of 500 bps over a 10-meter free space with a mean quantum bit error rate of 0.49%$\pm$0.27% is achieved. The security is analyzed under the scenario that Eve performs collective attack and photon number splitting collective attack. Our results show that quantum secure direct communication is feasible in free space.

quant-ph↗

arXiv2020

Similarity and delay between two non-narrow-band time signals

Correlation coefficient is usually used to measure the correlation degree between two time signals. However, its performance will drop or even fail if the signals are noised. Based on the time-frequency phase spectrum (TFPS) provided by normal time-frequency transform (NTFT), similarity coefficient is proposed to measure the similarity between two non-narrow-band time signals, even if the signals are noised. The basic idea of the similarity coefficient is to translate the interest part of signal f1(t)'s TFPS along the time axis to couple with signal f2(t)'s TFPS. Such coupling would generate a maximum if f1(t)and f2(t) are really similar to each other in time-frequency structure. The maximum, if normalized, is called similarity coefficient. The location of the maximum indicates the time delay between f1(t) and f2(t). Numerical results show that the similarity coefficient is better than the correlation coefficient in measuring the correlation degree between two noised signals. Precision and accuracy of the time delay estimation (TDE) based on the similarity analysis are much better than those based on cross-correlation (CC) method and generalized CC (GCC) method under low SNR.

eess.SP↗

arXiv2019

An Adaptive Moving Finite Element Method for Steady Low Mach Number Compressible Combustion Problems

This work surveys an r-adaptive moving mesh finite element method for the numerical solution of premixed laminar flame problems. Since the model of chemically reacting flow involves many different modes with diverse length scales, the computation of such a problem is often extremely time-consuming. Importantly, to capture the significant characteristics of the flame structure when using detailed chemistry, a much more stringent requirement on the spatial resolution of the interior layers of some intermediate species is necessary. Here, we propose a moving mesh method in which the mesh is obtained from the solution of so-called moving mesh partial differential equations. Such equations result from the variational formulation of a minimization problem for a given target functional that characterizes the inherent difficulty in the numerical approximation of the underlying physical equations. Adaptive mesh movement has emerged as an area of intense research in mesh adaptation in the last decade. With this approach points are only allowed to be shifted in space leaving the topology of the grid unchanged. In contrast to methods with local refinement, data structure hence is unchanged and load balancing is not an issue as grid points remain on the processor where they are. We will demonstrate the high potential of moving mesh methods for effectively optimizing the distribution of grid points to reach the required resolution for chemically reacting flows with extremely thin boundary layers.

math.NA↗

arXiv2018

Implementation and Security Analysis of Practical Quantum Secure Direct Communication

Fast development of supercomputer and perspective quantum computer is posing increasing serious threats to communication security. Based on the laws of quantum mechanics, quantum communication offers provable security of communication, and is a promising solution to counter such threats. Quantum secure direct communication (QSDC) is one of the important branches of quantum communication. Different from other branches of quantum communication, it transmits secret information directly. Recently, remarkable progress has been made in the proof-of-principle experimental demonstrations of QSDC. However, it remains a technical feast to march QSDC into practical application. Here, we report an implementation of practical quantum secure communication system. The security is analyzed in the Wyner wiretap channel theory. The system uses a coding scheme based on concatenation of low density parity check (LDPC) codes, which works in a regime with realistic environment of high noise and high loss. The present system operates with a repetition rate of 1 MHz, and at a distance of 1.5 kilometers. The secure communication rate is 50 bps, which can effectively send text message and files such as image and sounds with a reasonable size.

quant-ph↗

arXiv2016

Defending against malicious peripherals with Cinch

Malicious peripherals designed to attack their host computers are a growing problem. Inexpensive and powerful peripherals that attach to plug-and-play buses have made such attacks easy to mount. Making matters worse, commodity operating systems lack coherent defenses, and users are often unaware of the scope of the problem. We present Cinch, a pragmatic response to this threat. Cinch uses virtualization to attach peripheral devices to a logically separate, untrusted machine, and includes an interposition layer between the untrusted machine and the protected one. This layer regulates interaction with devices according to user-configured policies. Cinch integrates with existing OSes, enforces policies that thwart real-world attacks, and has low overhead.

cs.OS↗