Author name cluster

Zhijun Li

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

17 papers

2 author rows

JAIR Journal 2026 Journal Article

TeamTTA: Efficient Multi-Device Collaboration for Open-Set Test-Time Adaptation via Cloud Integration

Anqi Lu
Youbing Hu
Yun Cheng
Dawei Wei
Zhiqiang Cao
Jie Liu
Zhijun Li

Deep neural networks (DNNs) deployed on edge devices often suffer from severe performance degradation when exposed to dynamic and continually shifting environments. Test-time adaptation (TTA) has emerged as a promising solution by updating models online with incoming test data. However, edge deployment poses unique challenges: limited computational resources, latency caused by adaptation delays, and knowledge isolation across devices. The situation becomes even more complex in open-world scenarios, where the presence of unknown categories further disrupts adaptation. To overcome these limitations, we propose TeamTTA, a cloud-integrated framework designed for efficient multi-device collaboration open-set test-time adaptation. Specifically, TeamTTA aggregates reliable samples from multiple edge devices through crowdsourcing, uploads them to the cloud, and maintains a memory buffer for continual adaptation. A large vision model (LVM) in the cloud leverages its zero-shot generalization ability to filter out open-set samples and acts as a teacher model, distilling its knowledge into a replicated student edge model stored in the cloud. The adapted model parameters, or alternatively global statistics under poor network conditions, are then transmitted back to the edge devices for efficient inference. Extensive experiments on standard public TTA benchmarks, including corrupted and open-set datasets, show that TeamTTA achieves superior adaptation accuracy, robustness to distribution shifts, and communication efficiency, outperforming state-of-the-art TTA baselines. These results validate the effectiveness of integrating cloud-edge collaboration and LVM-driven knowledge distillation for real-world edge intelligence.

PDF Details DOI

IROS Conference 2025 Conference Paper

PD-VLA: Accelerating Vision-Language-Action Model Integrated with Action Chunking via Parallel Decoding

Wenxuan Song
Jiayi Chen
Pengxiang Ding
Han Zhao 0008
Wei Zhao
Zhide Zhong
Zongyuan Ge
Zhijun Li

Vision-Language-Action (VLA) models demonstrate remarkable potential for generalizable robotic manipulation. The performance of VLA models can be improved by integrating with action chunking, a critical technique for effective control. However, action chunking linearly scales up action dimensions in VLA models with increased chunking sizes. This reduces the inference efficiency. Therefore, accelerating VLA integrated with action chunking is an urgent need. To tackle this problem, we propose PD-VLA, the first parallel decoding framework for VLA models integrated with action chunking. Our framework reformulates autoregressive decoding as a nonlinear system solved by parallel fixed-point iterations. This approach preserves model performance with mathematical guarantees while significantly improving decoding speed. In addition, it enables training-free acceleration without architectural changes, as well as seamless synergy with existing acceleration techniques. Extensive simulations validate that our PD-VLA maintains competitive success rates while achieving 2. 52× execution frequency on manipulators (with 7 degrees of freedom) compared with the fundamental VLA model. Furthermore, we experimentally identify the most effective settings for acceleration. Finally, real-world experiments validate its high applicability across different tasks.

Details

EAAI Journal 2025 Journal Article

Self-information and prediction mask enhanced blind inpainting network for dunhuang murals

Jiahao Meng
Weirong Liu
Changhong Shi
Zhijun Li
Jie Liu

Blind image inpainting methods based on deep learning have shown promising results in digital image inpainting of dunhuang mural images in recent years. However, current blind inpainting methods still suffer from color patches and structural confusion in the repair results caused by contamination of damaged features and sub-network interference. To address the above problems, a self-information and prediction mask enhanced blind inpainting network (SIME-BINet) for dunhuang mural images is proposed. SIME-BINet redesigns blind inpainting method of phased guidance paradigm into information enhance paradigm, which continuously optimizes enhanced information in dynamic form during training process and provides guidance for encoding process. Meanwhile, an information enhanced transformer block is designed to overcome the problem of damaged feature contamination by introducing enhanced information. Experiments show that SIME-BINet outperforms recent state-of-the-art blind inpainting methods on DhMurals1714 dataset and real damage mask. SIME-BINet offers a new paradigm for blind image inpainting based on deep learning and provides an innovative approach for inpainting of dunhuang mural images. The code, data, and pre-trained models will be made available at https: //github. com/IPCSRG/SIME-BINet after the paper is published.

Details DOI

EAAI Journal 2024 Journal Article

Degression receptive field network for image inpainting

Jiahao Meng
Weirong Liu
Changhong Shi
Zhijun Li
Chaorong Liu

—Multi-stage image inpainting methods from coarse-to-fine have achieved satisfactory inpainting results in recent years. However, an in-depth analysis of multi-stage inpainting networks reveals that simply increasing complexity of refined network may lead to degradation problems. The paper proposes a degression receptive field network (DRFNet) via multi-head attention mechanism and U-shaped network with different receptive fields to address above phenomenon that existing image inpainting methods have detail blur and artifacts due to insufficient constraints. Initially, DRFNet innovatively takes receptive field as a perspective and consists of five sub-networks with decreasing receptive fields. Secondly, an easy-to-use TransConv module is designed to overcome the problem of local-pixel influence in convolution. Experiments show that comprehensive optimal rate of DRFNet on L1 error, Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), Fréchet Inception Distance (FID), and Learned Perceptual Image Patch Similarity (LPIPS) is more than 82. 86% on all three benchmark datasets, which achieves state-of-the-art results. Moreover, real-world experiments demonstrate the potential of DRFNet for object removal and image editing. The code is available at: https: //github. com/IPCSRG/DRFNet-Inpainting. git.

Details DOI

IJCAI Conference 2024 Conference Paper

ELF-UA: Efficient Label-Free User Adaptation in Gaze Estimation

Yong Wu
Yang Wang
Sanqing Qu
Zhijun Li
Guang Chen

We consider the problem of user-adaptive 3D gaze estimation. The performance of person-independent gaze estimation is limited due to interpersonal anatomical differences. Our goal is to provide a personalized gaze estimation model specifically adapted to a target user. Previous work on user-adaptive gaze estimation requires some labeled images of the target person data to fine-tune the model at test time. However, this can be unrealistic in real-world applications, since it is cumbersome for an end-user to provide labeled images. In addition, previous work requires the training data to have both gaze labels and person IDs. This data requirement makes it infeasible to use some of the available data. To tackle these challenges, this paper proposes a new problem called efficient label-free user adaptation in gaze estimation. Our model only needs a few unlabeled images of a target user for the model adaptation. During offline training, we have some labeled source data without person IDs and some unlabeled person-specific data. Our proposed method uses a meta-learning approach to learn how to adapt to a new user with only a few unlabeled images. Our key technical innovation is to use a generalization bound from domain adaptation to define the loss function in meta-learning, so that our method can effectively make use of both the labeled source data and the unlabeled person-specific data during training. Extensive experiments validate the effectiveness of our method on several challenging benchmarks.

PDF Details DOI

AAAI Conference 2024 Conference Paper

LF-ViT: Reducing Spatial Redundancy in Vision Transformer for Efficient Image Recognition

Youbing Hu
Yun Cheng
Anqi Lu
Zhiqiang Cao
Dawei Wei
Jie Liu
Zhijun Li

The Vision Transformer (ViT) excels in accuracy when handling high-resolution images, yet it confronts the challenge of significant spatial redundancy, leading to increased computational and memory requirements. To address this, we present the Localization and Focus Vision Transformer (LF-ViT). This model operates by strategically curtailing computational demands without impinging on performance. In the Localization phase, a reduced-resolution image is processed; if a definitive prediction remains elusive, our pioneering Neighborhood Global Class Attention (NGCA) mechanism is triggered, effectively identifying and spotlighting class-discriminative regions based on initial findings. Subsequently, in the Focus phase, this designated region is used from the original image to enhance recognition. Uniquely, LF-ViT employs consistent parameters across both phases, ensuring seamless end-to-end optimization. Our empirical tests affirm LF-ViT's prowess: it remarkably decreases Deit-S's FLOPs by 63% and concurrently amplifies throughput twofold. Code of this project is at https://github.com/edgeai1/LF-ViT.git.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Structured Matrix Basis for Multivariate Time Series Forecasting with Interpretable Dynamics

Xiaodan Chen
Xiucheng Li
Xinyang Chen
Zhijun Li

Multivariate time series forecasting is of central importance in modern intelligent decision systems. The dynamics of multivariate time series are jointly characterized by temporal dependencies and spatial correlations. Hence, it is equally important to build the forecasting models from both perspectives. The real-world multivariate time series data often presents spatial correlations that show structures and evolve dynamically. To capture such dynamic spatial structures, the existing forecasting approaches often rely on a two-stage learning process (learning dynamic series representations and then generating spatial structures), which is sensitive to the small time-window input data and has high variance. To address this, we propose a novel forecasting model with a structured matrix basis. At its core is a dynamic spatial structure generation function whose output space is well-constrained and the generated structures have lower variance, meanwhile, it is more expressive and can offer interpretable dynamics. This is achieved via a novel structured parameterization and imposing structure regularization on the matrix basis. The resulting forecasting model can achieve up to $8. 5\%$ improvements over the existing methods on six benchmark datasets, and meanwhile, it enables us to gain insights into the dynamics of underlying systems.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

VOCE: Variational Optimization with Conservative Estimation for Offline Safe Reinforcement Learning

Jiayi Guan
Guang Chen
Jiaming Ji
Long Yang
Ao Zhou
Zhijun Li
Changjun Jiang

Offline safe reinforcement learning (RL) algorithms promise to learn policies that satisfy safety constraints directly in offline datasets without interacting with the environment. This arrangement is particularly important in scenarios with high sampling costs and potential dangers, such as autonomous driving and robotics. However, the influence of safety constraints and out-of-distribution (OOD) actions have made it challenging for previous methods to achieve high reward returns while ensuring safety. In this work, we propose a Variational Optimization with Conservative Eestimation algorithm (VOCE) to solve the problem of optimizing safety policies in the offline dataset. Concretely, we reframe the problem of offline safe RL using probabilistic inference, which introduces variational distributions to make the optimization of policies more flexible. Subsequently, we utilize pessimistic estimation methods to estimate the Q-value of cost and reward, which mitigates the extrapolation errors induced by OOD actions. Finally, extensive experiments demonstrate that the VOCE algorithm achieves competitive performance across multiple experimental tasks, particularly outperforming state-of-the-art algorithms in terms of safety.

PDF Details

EAAI Journal 2022 Journal Article

New adaptive robust H ∞ control of smart structures using synchrosqueezed wavelet transform and recursive least-squares algorithm

Zhijun Li
Hojjat Adeli

Two kinds of uncertainties, one due to the dynamic earthquake loads with a wide frequency band and the other due to structural parameters exist in large and complex real-life structures. Most existing control algorithms consider only one of them, resulting in difficulty to guarantee necessary control performance for large complex structures, such as better vibration suppression on structural peak response and robustness performance. Considering the two uncertainties simultaneously, in this paper, a new adaptive robust H ∞ control methodology is presented for vibration control of structures through adroit integration of synchrosqueezed wavelet transform (SWT) and recursive least-squares (RLS) algorithm. The robust H ∞ control is more effective than the traditional LQR/LQG control in terms of the stability and robustness of the control system. The external excitation signal from ground sensors is filtered by a low-pass filter based on SWT and then inputted into the filtered-x RLS adaptive controller. The effectiveness, accuracy, and computational efficiency of the new adaptive control method is demonstrated using a 76-story wind-excited benchmark super high-rise building structure and a 24-story shear-wall building with an active tuned mass damper (ATMD) system on the top floor. Compared with the existing linear quadratic Gaussian control algorithm, the wavelet-hybrid feedback-least mean square algorithm and the robust H ∞ control algorithm, the simulation results show that the control effect and robust performance indexes of the structure are increased by 5%–35% and 5%–25%, respectively, using the new control methodology.

Details DOI

AAAI Conference 2021 Conference Paper

PointINet: Point Cloud Frame Interpolation Network

Fan Lu
Guang Chen
Sanqing Qu
Zhijun Li
Yinlong Liu
Alois Knoll

LiDAR point cloud streams are usually sparse in time dimension, which is limited by hardware performance. Generally, the frame rates of mechanical LiDAR sensors are 10 to 20 Hz, which is much lower than other commonly used sensors like cameras. To overcome the temporal limitations of LiDAR sensors, a novel task named Point Cloud Frame Interpolation is studied in this paper. Given two consecutive point cloud frames, Point Cloud Frame Interpolation aims to generate intermediate frame(s) between them. To achieve that, we propose a novel framework, namely Point Cloud Frame Interpolation Network (PointINet). Based on the proposed method, the low frame rate point cloud streams can be upsampled to higher frame rates. We start by estimating bi-directional 3D scene flow between the two point clouds and then warp them to the given time step based on the 3D scene flow. To fuse the two warped frames and generate intermediate point cloud(s), we propose a novel learning-based points fusion module, which simultaneously takes two warped point clouds into consideration. We design both quantitative and qualitative experiments to evaluate the performance of the point cloud frame interpolation method and extensive experiments on two large scale outdoor LiDAR datasets demonstrate the effectiveness of the proposed PointINet. Our code is available at https: //github. com/ispc-lab/PointINet. git.

PDF Details

EAAI Journal 2016 Journal Article

New discrete-time robust H2/H∞ algorithm for vibration control of smart structures using linear matrix inequalities

Zhijun Li
Hojjat Adeli

In real structural systems, such as a building structure or a mechanical system, due to inherent structural modeling approximations and errors, and changeable and unpredictable environmental loads, the structural response unavoidably involves uncertainties. These uncertainties can reduce the performance of a control algorithm significantly and possibly make it unstable. In this paper, based on the theories of the Bounded Real Lemma and the linear matrix inequalities (LMI), a novel discrete-time robust H 2/H ∞ control algorithm is presented which not only reduces the structural peak response caused by external dynamic forces but also is robust and stable in the presence of parametric uncertainties which is always the case in real-life structures. To facilitate practical implementation, the uncertainties of structural parameters are considered in the time domain as opposed to the frequency domain. Compared with traditional H ∞ control methods, the new control algorithm proposes a convenient design procedure to facilitate practical implementations of active control of complex and large structural systems through the use of a quadratic performance index and the LMI-based solution method. The effectiveness of the new discrete-time robust H 2/H ∞ adaptive control algorithm is demonstrated using a three-story frame with active bracing systems (ABS) and a ten-story frame with an active tuned mass damper (ATMD).

Details DOI

EAAI Journal 2014 Journal Article

Adaptive fuzzy-based motion generation and control of mobile under-actuated manipulators

Zhijun Li
Chenguang Yang
Chun-Yi Su
Wenjun Ye

In this paper, adaptive fuzzy-based motion generation and control are investigated for nonholonomic mobile manipulators with an under-actuated dyanmics model, in the presence of parametric and functional uncertainties. It is well known that the constraints of this kind of system consist of kinematic constraints for the mobile platform and dynamic constraints for the under-actuated manipulator with a passive joint. Through using dynamic coupling property of nonholonomic mobile manipulators, we can decouple the dynamics into a fully actuated subsystem and an unactuated subsystem. Then adaptive control is employed for the fully actuated subsystem using fuzzy logic approximation. Since the non-actuated subsystem cannot be directly manipulated by torque inputs but can be indirectly affected by the motion of the actuated subsystem, the reference trajectory of the actuated subsystem is planned by the fuzzy logic system based motion generator. Rigorous theoretic analysis has been established to show that the proposed trajectory generation and control are able to achieve dynamic stability, motion tracking and optimized dynamics. Simulation studies have further validated the efficiency of the developed scheme.

Details DOI

JBHI Journal 2014 Journal Article

sEMG-Based Joint Force Control for an Upper-Limb Power-Assist Exoskeleton Robot

Zhijun Li
Baocheng Wang
Fuchun Sun
Chenguang Yang
Qing Xie
Weidong Zhang

This paper investigates two surface electromyogram (sEMG)-based control strategies developed for a power-assist exoskeleton arm. Different from most of the existing position control approaches, this paper develops force control methods to make the exoskeleton robot behave like humans in order to provide better assistance. The exoskeleton robot is directly attached to a user’s body and activated by the sEMG signals of the user’s muscles, which reflect the user’s motion intention. In the first proposed control method, the forces of agonist and antagonist muscles pair are estimated, and their difference is used to produce the torque of the corresponding joints. In the second method, linear discriminant analysis-based classifiers are introduced as the indicator of the motion type of the joints. Then, the classifier’s outputs together with the estimated force of corresponding active muscle determine the torque control signals. Different from the conventional approaches, one classifier is assigned to each joint, which decreases the training time and largely simplifies the recognition process. Finally, the extensive experiments are conducted to illustrate the effectiveness of the proposed approaches.

Details DOI

JBHI Journal 2013 Journal Article

Boosting-Based EMG Patterns Classification Scheme for Robustness Enhancement

Zhijun Li
Baocheng Wang
Chenguang Yang
Qing Xie
Chun-Yi Su

The high conventional accuracy of pattern recognition-based surface myoelectric classification in laboratory experiments does not necessarily result in high accessibility to practical protheses. An obvious reason is the effect of signals of untrained classes caused by the relatively small training dataset. In order to make the classifier robust to untrained classes, a classification scheme is developed based on boosting and random forest classifiers in this paper. Meanwhile, a threshold, the post probability of the prediction, is introduced as a balance (i. e. , adjust) between the accurate classification and the rejection of the samples belonging to some untrained classes. The experiments are conducted to compare with other two schemes using linear discriminant analysis and support vector machines. Surface electromyogram signals, labeled with seven isometric movements, are collected from six healthy subjects' forearm. It is shown that the proposed scheme can reach up to about 92% accuracy in recognizing trained classes and 20% for untrained classes. Through adjusting the threshold, the accuracy of rejecting untrained classes reaches up to around 80%, with small decrease in recognizing trained classes (down to 80%). In the analysis of experiments' results, we also find that the proposed scheme has better error distribution among the classes.

Details DOI

EAAI Journal 2012 Journal Article

Freely-drawn sketches interpretation using SVMs-chain modeling

Kun Yang
Zhijun Li
Jingwei Ye

The growing popularity of tablet PCs and intelligent pen-centric computing has increased the importance of freehand sketch recognition algorithms. In this paper, the proposed method integrates the temporal, spatial and geometric constraint information to improve the recognition accuracy. To interpret the sketch as an incremental process, the paper investigates the use of the information fusion technique with Support Vector Machines (SVMs) chain for modeling and understanding the spatial and temporal information of sketch sequences. Online sketch recognition is achieved through the use of the SVMs-chain for systematically modeling the dynamic and stochastic behaviors of the sketch. To validate its efficiency, the experimental results in various domains and the comparison with traditional Hidden Markov Models have been presented.

Details DOI

ICRA Conference 2011 Conference Paper

Adaptive fuzzy control for synchronization of nonlinear teleoperators with stochastic time-varying communication delays

Zhijun Li
Nan Ding
Xiaoqing Cao

In this paper, adaptive fuzzy control is investigated for nonlinear teleoperators with time-delays, which ensures synchronization of positions and velocities of the master and slave manipulators, and does not rely on the use of the scattering transformation. Compared with previous passivity framework, the communication delays are assumed to be stochastic time varying. By feedback linearization, the nonlinear dynamics of the tele-operation system is transformed into two sub systems: local master/slave position control with unmodelled dynamics and delayed motion synchronization. Then, based on linear matrix inequalities (LMI) and Markov jump linear systems, adaptive fuzzy control strategies are developed for the nonlinear teleoperators with time-delay, modeling uncertainties and external disturbances by using the approximation property of the fuzzy logic systems. It is proven that the master-slave tele operation system is stochastically stable in mean square under specific LMI conditions, and all the signals of the resulting closed-loop system are uniformly bounded.

Details

EAAI Journal 2008 Journal Article

Adaptive neural-fuzzy control of uncertain constrained multiple coordinated nonholonomic mobile manipulators

Zhijun Li
Weidong Chen

Most studies on the coordination of multiple mobile manipulators system assume exact knowledge of system kinematics and dynamics, and deal only with motion tracking control. However, actual applications may involve tasks in which multiple coordinated mobile manipulators system is required to keep contact on the contour of the constraint surface in tasks. In this paper, we consider multiple mobile manipulators grasping a rigid object in contact with deformable working surfaces, whose geometric and physical model is unknown. The contact forces are nonlinear and unknown. Adaptive neuro-fuzzy (NF) control for coordinated mobile manipulators is proposed for robust force/motion tracking on the constraint surface while it is in motion. The control law is decoupled in three subspaces and adaptive tuning mechanism is developed to deal with the uncertain environmental constraints, disturbances, and unknown robotic dynamics. The proposed adaptive NF hybrid force/motion controller guarantees robust tracking of the desired motion and force trajectories. Simulation examples are presented to illustrate the results.

Details DOI