Li Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Li Zhang has a total of 72 co-authors. Here are the top co-authors by number of shared publications:
Jieqing Tan
5 shared publications- Jingao Xu
The University of Hong Kong
5 shared publications - Zheng Yang
Tsinghua University
5 shared publications - Danyang Li
School of Software and BNRist, Tsinghua University, Beijing, China
4 shared publications - Jieqing Tan
School of Mathematics Hefei University of Technology Hefei China
4 shared publications - Xianyu Ge
School of Electronic and Information Engineering, Anhui Jianzhu University, Hefei, China
4 shared publications
Latest Publications
- 11 September 2026
- Published by Springer Nature
- p. 110-129
Towards Robust Driving Perception: A Flexible Scale-Driven Family for Self-Supervised Monocular Depth Estimation
- 11 September 2026
- Published by Springer Nature
- p. 110-129
Towards Robust Driving Perception: A Flexible Scale-Driven Family for Self-Supervised Monocular Depth Estimation
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (8) , 9859-9866
MixVIO: Selective Visual-Inertial Odometry via Phase-Aware Attention and Hybrid Kernel Modeling
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (8) , 9859-9866
MixVIO: Selective Visual-Inertial Odometry via Phase-Aware Attention and Hybrid Kernel Modeling
- 31 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (6) , 6839-6846
VistaDepth: Improving Far-Range Depth Estimation With Spectral Modulation and Adaptive Reweighting
Monocular depth estimation infers per-pixel depth from a single RGB image. It remains particularly challenging in far-range regions, where sparse observations and long-tailed depth distributions bias learning toward near-range content. Diffusion models offer a promising alternative to discriminative foundation models by leveraging rich generative priors for zero-shot generalization. However, existing methods still struggle to recover fine distant structures under weak supervision and limited image evidence. To address this, we propose VistaDepth, a diffusion framework for far-range depth estimation. It improves far-range prediction through two complementary mechanisms: Latent Frequency Modulation (LFM), which refines latent features via dynamic, content-aware spectral filtering, and BiasMap, which adaptively reweights the diffusion objective toward distant, structurally informative regions while remaining aligned with denoising. Extensive experiments on five benchmarks show that VistaDepth achieves state-of-the-art performance among diffusion-based methods. Despite training on orders of magnitude less data than discriminative baselines, it remains competitive globally and excels in challenging far-range regions.
- 31 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (6) , 6839-6846
VistaDepth: Improving Far-Range Depth Estimation With Spectral Modulation and Adaptive Reweighting
Monocular depth estimation infers per-pixel depth from a single RGB image. It remains particularly challenging in far-range regions, where sparse observations and long-tailed depth distributions bias learning toward near-range content. Diffusion models offer a promising alternative to discriminative foundation models by leveraging rich generative priors for zero-shot generalization. However, existing methods still struggle to recover fine distant structures under weak supervision and limited image evidence. To address this, we propose VistaDepth, a diffusion framework for far-range depth estimation. It improves far-range prediction through two complementary mechanisms: Latent Frequency Modulation (LFM), which refines latent features via dynamic, content-aware spectral filtering, and BiasMap, which adaptively reweights the diffusion objective toward distant, structurally informative regions while remaining aligned with denoising. Extensive experiments on five benchmarks show that VistaDepth achieves state-of-the-art performance among diffusion-based methods. Despite training on orders of magnitude less data than discriminative baselines, it remains competitive globally and excels in challenging far-range regions.
- 3 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-10
EVIU-Link: Composite Collaborative Perception Framework for Enhanced Indoor Localization
Visual-inertial odometry (VIO) is widely used for mobile robot localization in indoor areas. Recent studies show that integrating global sensors such as ultra-wideband (UWB) into VIO can improve long-term robustness, but existing multi-sensor fusion schemes still face challenges. First, visual feature tracking, a core component of VIO, is easily degraded under low-texture, low-light, or fast-motion conditions, where current visual enhancement methods remain ineffective. Second, most fusion strategies are constrained to a single coupling paradigm (loose/tight) and a single estimation scheme (filter/optimization), which makes it difficult to jointly balance accuracy, efficiency, and robustness. To address these issues, we propose EVIU-link, an event-based VIO-UWB fusion framework. The key innovations are twofold: (1) an event-enhanced vision module, where feature-density monitoring dynamically triggers event-flow tracking to maintain robust feature matching under challenging visual conditions; and (2) a hierarchical fusion strategy that performs local pose estimation within a VIO-UWB constrained factor graph, while simultaneously applying UWB ranging constraints at the global level to correct accumulated drift. Experiments on public datasets and real-world indoor scenarios demonstrate that EVIU-link achieves superior performance in challenging environments, indicating its promising application potential.
- 3 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-10
EVIU-Link: Composite Collaborative Perception Framework for Enhanced Indoor Localization
Visual-inertial odometry (VIO) is widely used for mobile robot localization in indoor areas. Recent studies show that integrating global sensors such as ultra-wideband (UWB) into VIO can improve long-term robustness, but existing multi-sensor fusion schemes still face challenges. First, visual feature tracking, a core component of VIO, is easily degraded under low-texture, low-light, or fast-motion conditions, where current visual enhancement methods remain ineffective. Second, most fusion strategies are constrained to a single coupling paradigm (loose/tight) and a single estimation scheme (filter/optimization), which makes it difficult to jointly balance accuracy, efficiency, and robustness. To address these issues, we propose EVIU-link, an event-based VIO-UWB fusion framework. The key innovations are twofold: (1) an event-enhanced vision module, where feature-density monitoring dynamically triggers event-flow tracking to maintain robust feature matching under challenging visual conditions; and (2) a hierarchical fusion strategy that performs local pose estimation within a VIO-UWB constrained factor graph, while simultaneously applying UWB ranging constraints at the global level to correct accumulated drift. Experiments on public datasets and real-world indoor scenarios demonstrate that EVIU-link achieves superior performance in challenging environments, indicating its promising application potential.
- 2 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (2) , 1870-1884
Secure Charging Scheduling in Wireless Rechargeable Sensor Networks
Wireless Rechargeable Sensor Networks (WRSNs) promise to address the limited energy resource issue for sensor nodes through wireless power transfer technology. However, WRSNs are vulnerable to various security threats, such as compromised node attack and malicious mobile charger (MC) attack, which can disrupt the charging process and degrade charging efficiency. In this work, we investigate the eneRgy conversion Efficiency maximization problem unDer chargIng attackS (REDIS). We propose a blockchain-based framework that employs a lightweight multi-layer storage approach tailored for resource-constrained sensor nodes and features consensus algorithms that validate charging transactions. Furthermore, we introduce a validation node selection strategy that integrates consensus execution with charging scheduling, reducing energy consumption, and improving energy efficiency. Extensive simulations and experiments validate the effectiveness of our framework, improving energy efficiency by 30% and as much as 5 times in networks without attacks and those under full attacks, respectively.
- 2 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (2) , 1870-1884
Secure Charging Scheduling in Wireless Rechargeable Sensor Networks
Wireless Rechargeable Sensor Networks (WRSNs) promise to address the limited energy resource issue for sensor nodes through wireless power transfer technology. However, WRSNs are vulnerable to various security threats, such as compromised node attack and malicious mobile charger (MC) attack, which can disrupt the charging process and degrade charging efficiency. In this work, we investigate the eneRgy conversion Efficiency maximization problem unDer chargIng attackS (REDIS). We propose a blockchain-based framework that employs a lightweight multi-layer storage approach tailored for resource-constrained sensor nodes and features consensus algorithms that validate charging transactions. Furthermore, we introduce a validation node selection strategy that integrates consensus execution with charging scheduling, reducing energy consumption, and improving energy efficiency. Extensive simulations and experiments validate the effectiveness of our framework, improving energy efficiency by 30% and as much as 5 times in networks without attacks and those under full attacks, respectively.
- 27 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (19) , 36744-36753
IPFNet: Interactive Perceptive Fusion Network for Self-Supervised Monocular Depth Estimation
Monocular depth estimation (MDE) is an essential challenge in computer vision, and deep learning has provided numerous solutions. Despite advancements, deep learning models frequently encounter limitations, including the ambiguity in feature representation, particularly around object boundaries and fine structures, as well as the dependency on large-scale, accurately labeled datasets. The latter are costly and difficult to acquire, especially in outdoor and dynamic environments. Building upon recent research, we introduce a novel method for an interactive perception fusion network(IPFNet) framework that uniquely leverages semantic and inertial measurement unit(IMU) cues in complex environments. Our approach emphasizes the precise fusion of semantic and depth features particularly at object boundaries, proposing the symmetric extraction propagation(SEP) that facilitates a deep interaction of rich contextual information across multiple scales between semantic and geometric domains. Meanwhile, we introduce a visual-inertial fusion filter, which jointly models the IMU-based motion propagation and a deep learning-based pose estimation network. The resulting posterior pose is employed to construct an IMU-guided photometric consistency loss. By fusing multimodal data, this module enriches the diversity of supervision signals and effectively mitigates the limitations of monocular vision in handling illumination changes, occlusions, and dynamic objects. Our model undergoes extensive evaluation on established benchmark datasets, with comparisons to existing state-of-the-art (SOTA) methods. The results indicate enhanced depth estimation accuracy, improved model robustness, and better performance under challenging lighting conditions.
- 27 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (19) , 36744-36753
IPFNet: Interactive Perceptive Fusion Network for Self-Supervised Monocular Depth Estimation
Monocular depth estimation (MDE) is an essential challenge in computer vision, and deep learning has provided numerous solutions. Despite advancements, deep learning models frequently encounter limitations, including the ambiguity in feature representation, particularly around object boundaries and fine structures, as well as the dependency on large-scale, accurately labeled datasets. The latter are costly and difficult to acquire, especially in outdoor and dynamic environments. Building upon recent research, we introduce a novel method for an interactive perception fusion network(IPFNet) framework that uniquely leverages semantic and inertial measurement unit(IMU) cues in complex environments. Our approach emphasizes the precise fusion of semantic and depth features particularly at object boundaries, proposing the symmetric extraction propagation(SEP) that facilitates a deep interaction of rich contextual information across multiple scales between semantic and geometric domains. Meanwhile, we introduce a visual-inertial fusion filter, which jointly models the IMU-based motion propagation and a deep learning-based pose estimation network. The resulting posterior pose is employed to construct an IMU-guided photometric consistency loss. By fusing multimodal data, this module enriches the diversity of supervision signals and effectively mitigates the limitations of monocular vision in handling illumination changes, occlusions, and dynamic objects. Our model undergoes extensive evaluation on established benchmark datasets, with comparisons to existing state-of-the-art (SOTA) methods. The results indicate enhanced depth estimation accuracy, improved model robustness, and better performance under challenging lighting conditions.
- 26 July 2025
- Published by Springer Nature
- p. 145-155
Optimizing Landmark Graphs in DHRL: A Dual Approach of Attention and Weighted Sampling
- 26 July 2025
- Published by Springer Nature
- p. 145-155
Optimizing Landmark Graphs in DHRL: A Dual Approach of Attention and Weighted Sampling
- 20 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (13) , 24960-24969
DTDepth: A Self-Supervised Framework With Ultrasparse LiDAR for Image Depth Estimation
Self-supervised learning provides a promising approach for monocular depth estimation (MDE) tasks by mitigating the reliance on training data that is both scarce and expensive to acquire. However, prevailing image-based and multi-modal self-supervised MDE methods often fail to attain the high accuracy demanded by downstream applications. This study explores a self-supervised method leveraging the guidance from ultra-sparse LiDAR, to enable an effective and cost-efficient framework for dense depth prediction. Inspired by the convolutional neural network (CNN) and Vision Transformer (ViT) models, we propose DTDepth, a novel hybrid framework based on dilated convolution and transformer operators. Given the image and low-beam LiDAR as input, we designed an efficient fusion module in our encoder for the deep interaction between the dense image and sparse LiDAR features, then we extensively utilized attention mechanisms to enrich the aggregated features in the decoder. The key part of our framework is Joint Transformer and Dilated Convolution (JTDC) block, which can effectively consider global and local feature extraction. Departing from traditional dense depth completion, our innovative approach employs affordable low-beam LiDAR to guide the image depth prediction. We evaluated the performance of our model on KITTI and Make3D datasets, where we demonstrated that our proposed method significantly outperforms many existing state-of-the-art algorithms and narrows the performance gap between supervised and self-supervised methods.
- 20 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (13) , 24960-24969
DTDepth: A Self-Supervised Framework With Ultrasparse LiDAR for Image Depth Estimation
Self-supervised learning provides a promising approach for monocular depth estimation (MDE) tasks by mitigating the reliance on training data that is both scarce and expensive to acquire. However, prevailing image-based and multi-modal self-supervised MDE methods often fail to attain the high accuracy demanded by downstream applications. This study explores a self-supervised method leveraging the guidance from ultra-sparse LiDAR, to enable an effective and cost-efficient framework for dense depth prediction. Inspired by the convolutional neural network (CNN) and Vision Transformer (ViT) models, we propose DTDepth, a novel hybrid framework based on dilated convolution and transformer operators. Given the image and low-beam LiDAR as input, we designed an efficient fusion module in our encoder for the deep interaction between the dense image and sparse LiDAR features, then we extensively utilized attention mechanisms to enrich the aggregated features in the decoder. The key part of our framework is Joint Transformer and Dilated Convolution (JTDC) block, which can effectively consider global and local feature extraction. Departing from traditional dense depth completion, our innovative approach employs affordable low-beam LiDAR to guide the image depth prediction. We evaluated the performance of our model on KITTI and Make3D datasets, where we demonstrated that our proposed method significantly outperforms many existing state-of-the-art algorithms and narrows the performance gap between supervised and self-supervised methods.
- Open Access
- 8 July 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Sensor Networks
- Vol. 20 (4) , 1-22
HCCNet: Hybrid Coupled Cooperative Network for Robust Indoor Localization
Accurate localization of unmanned aerial vehicle (UAV) is critical for navigation in GPS-denied regions, which remains a highly challenging topic in recent research. This article describes a novel approach to multi-sensor hybrid coupled cooperative localization network (HCCNet) system that combines multiple types of sensors including camera, ultra-wideband (UWB), and inertial measurement unit (IMU) to address this challenge. The camera and IMU can automatically determine the position of UAV based on the perception of surrounding environments and their own measurement data. The UWB node and the UWB wireless sensor network (WSN) in indoor environments jointly determine the global position of UAV, and the proposed dynamic random sample consensus (D-RANSAC) algorithm can optimize UWB localization accuracy. To fully exploit UWB localization results, we provide an HCCNet system which combines the local pose estimator of visual inertial odometry (VIO) system with global constraints from UWB localization results. Experimental results show that the proposed D-RANSAC algorithm can achieve better accuracy than other UWB-based algorithms. The effectiveness of the proposed HCCNet method is verified by a mobile robot in real world and some simulation experiments in indoor environments.
- 8 July 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Sensor Networks
- Vol. 20 (4) , 1-22
Open AccessHCCNet: Hybrid Coupled Cooperative Network for Robust Indoor Localization
Accurate localization of unmanned aerial vehicle (UAV) is critical for navigation in GPS-denied regions, which remains a highly challenging topic in recent research. This article describes a novel approach to multi-sensor hybrid coupled cooperative localization network (HCCNet) system that combines multiple types of sensors including camera, ultra-wideband (UWB), and inertial measurement unit (IMU) to address this challenge. The camera and IMU can automatically determine the position of UAV based on the perception of surrounding environments and their own measurement data. The UWB node and the UWB wireless sensor network (WSN) in indoor environments jointly determine the global position of UAV, and the proposed dynamic random sample consensus (D-RANSAC) algorithm can optimize UWB localization accuracy. To fully exploit UWB localization results, we provide an HCCNet system which combines the local pose estimator of visual inertial odometry (VIO) system with global constraints from UWB localization results. Experimental results show that the proposed D-RANSAC algorithm can achieve better accuracy than other UWB-based algorithms. The effectiveness of the proposed HCCNet method is verified by a mobile robot in real world and some simulation experiments in indoor environments.
- 20 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-13
REFLoc: A Resilient Evolutionary Fusion Framework for Robust Indoor Localization
Location awareness is becoming increasingly crucial in unmanned aerial vehicles (UAVs). However, in the absence of global positioning system (GPS) signals in indoor scenes, how to obtain accurate positioning results has become a frontier issue that has attracted much attention. This paper proposes a resilient evolutionary multisensor fusion localization framework named “REFLoc” to achieve accurate indoor positioning of UAVs. Motivated by the ease of deployment and the wide feasibility of precise indoor localization, we introduce ultra-wideband (UWB) as global localization. The global localization provides the absolute position information of the UAV indoors, but it may be affected by signal blocking, occlusion, or interference. To address these noise effects, we introduce a hybrid threshold wavelet denoising (HTWD) algorithm to effectively preprocess UWB data. We also integrate visual inertial odometry (VIO) as the support of local tracking to provide continuous and high-frequency motion estimation. We dynamically adjust the weight parameters and noise covariance matrices by using the resilient evolutionary fusion method. Then we further seamlessly integrate the global localization and local tracking results in the optimized EKF framework. This improves the robustness of the system to environmental changes and outliers. We evaluate the performance of REFLoc in public datasets and in real world, and the experimental results show that the proposed framework can effectively improve the robustness and accuracy of the UAVs’ localization.
- 20 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-13
REFLoc: A Resilient Evolutionary Fusion Framework for Robust Indoor Localization
Location awareness is becoming increasingly crucial in unmanned aerial vehicles (UAVs). However, in the absence of global positioning system (GPS) signals in indoor scenes, how to obtain accurate positioning results has become a frontier issue that has attracted much attention. This paper proposes a resilient evolutionary multisensor fusion localization framework named “REFLoc” to achieve accurate indoor positioning of UAVs. Motivated by the ease of deployment and the wide feasibility of precise indoor localization, we introduce ultra-wideband (UWB) as global localization. The global localization provides the absolute position information of the UAV indoors, but it may be affected by signal blocking, occlusion, or interference. To address these noise effects, we introduce a hybrid threshold wavelet denoising (HTWD) algorithm to effectively preprocess UWB data. We also integrate visual inertial odometry (VIO) as the support of local tracking to provide continuous and high-frequency motion estimation. We dynamically adjust the weight parameters and noise covariance matrices by using the resilient evolutionary fusion method. Then we further seamlessly integrate the global localization and local tracking results in the optimized EKF framework. This improves the robustness of the system to environmental changes and outliers. We evaluate the performance of REFLoc in public datasets and in real world, and the experimental results show that the proposed framework can effectively improve the robustness and accuracy of the UAVs’ localization.