Mengxi Jiang

Dr. Mengxi Jiang

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 November 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
TWtiange wang
tiange wang
Guangsong YangGuangsong Yang
Guangsong Yang
PCPenghui Chen
Penghui Chen
ZXZhenghua Xu
Zhenghua Xu
MJMengxi Jiang
Mengxi Jiang
QYQiubo Ye
Qiubo Ye
With the continuous development of communication technology, the wireless communication environment becomes more and more complex with various intentional and unintentional signals. Radio signals are modulated in different ways. The traditional radio modulation recognition technology cannot recognize the modulation modes accurately. Consequently, the communication system has embraced Deep Learning (DL) models as they can automatically recognize the modulation modes and have better accuracy. This paper systematically summarizes the related contents of radio Automatic Modulation Recognition (AMR) based on DL over the last seven years. First, we summarize the current research status of modulation recognition and the necessity of AMR research based on DL. Then, we review current radio AMR methods based on DL. In addition, we also propose a network model of AMR based on Convolutional Neural Network (CNN) and prove its effectiveness. Finally, we highlight existing challenges and research directions of radio AMR based on DL.

13 June 2025 in Applied Sciences
Verified
Journal Article
Open access
PDF file
RLRuiquan Lin
Ruiquan Lin
YJYunwei Ji
Yunwei Ji
WDWanting Ding
Wanting Ding
TWTIANXIANG WU
TIANXIANG WU
YZYaosheng Zhu
Yaosheng Zhu
MJMengxi Jiang
Mengxi Jiang
Three-dimensional CAD reconstruction is a long-standing and important task in fields such as industrial manufacturing, architecture, medicine, film and television, research, and education. Reconstructing CAD models remains a persistent challenge in machine learning. There have been many studies on deep learning in the field of 3D reconstruction. In recent years, with the release of CAD datasets, there have been more and more studies on 3D CAD reconstruction using deep learning. With the continuous deepening of research, deep learning has significantly improved the performance of tasks in the field of CAD reconstruction. However, this task remains challenging due to data scarcity and labeling difficulties, model complexity, and lack of generality and adaptability. This paper reviews both classic and recent research results on 3D CAD reconstruction tasks based on deep learning. To the best of our knowledge, this is the first investigation focusing on the CAD reconstruction task in the field of deep learning. Since there are relatively few studies related to 3D CAD reconstruction, we also investigate the reconstruction and generation of 2D CAD sketches. According to the different input data, we divide all investigations into the following categories: point cloud input to 3D CAD models, sketch input to 3D CAD models, other input to 3D CAD models, reconstruction and generation of 2D sketches, characterization of CAD data, CAD datasets, and related evaluation indicators. Commonly used datasets are outlined in our taxonomy. We provide a brief overview of the current research background, challenges, and recent results. Finally, future research directions are discussed.

18 March 2024 in IEEE Transactions on Dependable and Secure Computing
Verified
Journal Article
Closed access
MJMengxi Jiang
Mengxi Jiang
YSYulei Sui
Yulei Sui
YLYunqi Lei
Yunqi Lei
XXXiaofei Xie
Xiaofei Xie
CLCuihua Li
Cuihua Li
ITIvor Tsang
Ivor Tsang
Adversarial attack is a crucial step when evaluating the reliability and robustness of deep neural networks (DNNs) models. Most existing attack approaches apply an end-to-end gradient update strategy to generate adversarial examples for a classification or regression problem. However, few of them consider the non-differentiable DNN models (e.g., coordinate regression model) that prevent end-to-end backpropagation resulting in the failure of gradient calculation. In this paper, we present a new adversarial example generation approach for both untargeted and targeted attacks on coordinate regression models with non-differentiable operations. The novelty of our approach lies in a k-layer penetrating representation, on which we perturb the hidden feature distribution of the k-th layer through relational guidance to influence the final output, in which end-to-end backpropagation is not required. Rather than modifying a large portion of the pixels in an image, the proposed approach only modifies a very small set of the input pixels. These pixels are carefully and precisely selected by three correlations between the input pixels and hidden features of the k-th layer of a DNN, thus significantly reducing the adversarial perturbation on a clean image. We successfully apply the proposed approach to two different tasks (i.e., 2D and 3D human pose estimation) which are typical applications of the coordinate regression learning. The comprehensive experiments demonstrate that our approach achieves better performance while using much less adversarial perturbation on clean images.

27 October 2025 in IEEE Sensors Journal
Verified
Journal Article
Closed access
Pingping ChenPingping Chen
Pingping Chen
XHXiaopeng Hu
Xiaopeng Hu
BJBiao Jiang
Biao Jiang
YZYaosheng Zhu
Yaosheng Zhu
MJMengxi Jiang
Mengxi Jiang
It is important to estimate the 3D human shape from a single image, as it is utilized in virtual reality, tactile games and human-computer interaction. Most previous studies attempted to establish a mapping from the input image to 3D shape coordinate spaces using neural network optimization based on Euclidean distance. However, such a strategy fails to maintain the aligned 2D-3D surface of the human body because structural topological associations have not been effectively considered. We propose a surface-aligned 3D human shape estimation via the Gaussian curvature structure learning approach. Specifically, we use positive or negative Gaussian curvature to model the convex and concave appearance of the human body. Then, we use them to guide our 3D human shape estimation network, capturing the structural relationships between the vertices and the local surfaces of the human body. The optimization of our network is simultaneously processed via coordinate and curvature features in both Euclidean and Riemann spaces to maintain the 2D-3D consistency of the surface and vertex on the human body. Furthermore, the designed curvature completion module can obtain more 3D surface prior, which mitigates the performance degradation problem with more complex deformation of the human body in an outdoor scene. Extensive testing on some selected representative benchmarks demonstrates the efficacy of our approach. Especially, it shows superiority in the detailed texture reconstruction.


15 December 2025 in IEEE Sensors Journal
Verified
Journal Article
Closed access
HCHonghui Chen
Honghui Chen
XPXinyang Piao
Xinyang Piao
SYSuo You
Suo You
MJMengxi Jiang
Mengxi Jiang
CCChengbin Chen
Chengbin Chen
Pingping ChenPingping Chen
Pingping Chen
Real-time shipping container code spotting plays a vital role in enabling intelligent inspection to optimize smart port logistics. Recent methods suffer from error accumulation and have complex pipelines, resulting in high computational complexity that limits the generalization ability for multi-directional, including horizontal, vertical, and multi-line (HVM) codes. Moreover, fixing the parameters of visual sensors (e.g., angle, position, and focal length) to avoid noise interference limits deployment on mobile devices. To address them, we propose an end-to-end framework for HVM code spotting. First, an inter-domain feature fusion network that includes the Self-Spatial Enhancement Module (SSEM) and Self-Channel Enhancement Module (SCEM) is proposed to reduce noise interference and improve the efficiency of the lightweight model. Next, a transformer-based branch with Character Contrastive Learning (CCL) loss is designed to enhance the representation of character features. Experimental results indicate that the proposed method achieves state-of-the-art performance, in which the F-measure and recognition accuracy reach 91.2% and 93.2%, respectively, in real-time.