Fei Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Fei Wang has a total of 49 co-authors. Here are the top co-authors by number of shared publications:
- Kun Li
United Arab Emirates University, Al Ain, United Arab Emirates
12 shared publications - Dan Guo
Hefei University of Technology, Hefei, China
8 shared publications - Yanyan Wei
Hefei University of Technology
7 shared publications - Meng Wang
Hefei University of Technology, Hefei, China
4 shared publications - Jihao Gu
University College London, London, United Kingdom
3 shared publications - Yuwei Wang
College of Engineering, Anhui Agricultural University, Hefei, China
3 shared publications
Latest Publications
- 12 August 2026
- Published by Springer Nature
- p. 60-76
Spatial-Temporal Decoupled Adapter for Micro-gesture Online Recognition
- 12 August 2026
- Published by Springer Nature
- p. 60-76
Spatial-Temporal Decoupled Adapter for Micro-gesture Online Recognition
- 12 August 2026
- Published by Springer Nature
- p. 17-27
Self-supervised Learning Matters: A Simple Ensemble Solution for Micro-gesture Recognition
- 12 August 2026
- Published by Springer Nature
- p. 17-27
Self-supervised Learning Matters: A Simple Ensemble Solution for Micro-gesture Recognition
- 12 August 2026
- Published by Springer Nature
- p. 104-121
Rethinking the Role of Feature Engineering and Learning Strategies in Few-Shot Hidden Emotion Recognition
- 12 August 2026
- Published by Springer Nature
- p. 104-121
Rethinking the Role of Feature Engineering and Learning Strategies in Few-Shot Hidden Emotion Recognition
- Open Access
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 9297-9308
XInsight: Integrative Stage-Consistent Psychological Counseling Support Agents for Digital Well-Being
Web-based platforms are becoming a primary channel for psychological support, yet most LLM-driven chatbots remain opaque, single-stage, and weakly grounded in established therapeutic practice. To address this gap, we present XInsight, a multi-agent framework that models psychological support as a stage-consistent workflow aligned with the classical Exploration-Insight-Action paradigm. Building on structured client representations, XInsight orchestrates specialized agents under a unified Reason-Intervene-Reflect cycle: an Exploration agent organizes background and concerns into a structured Case Conceptualization Form, a Routing agent performs Adaptive Therapeutic Routing (ATR) across SFBT, CBT, and MBCT, a unified Therapeutic agent executes school-consistent submodules, and a Consolidation agent guides review, skill integration, and relapse-prevention planning. A Recording agent continuously transforms open-ended web dialogues into standardized psychological artifacts, enhancing interpretability, continuity, and accountability. To support transparent assessment, we introduce XInsight-Bench with a Scale-Guided LLM Evaluation (SGLE) protocol that combines therapy-specific clinical scales with general counseling criteria. Experiments show improved paradigm alignment, multi-therapy integration, interaction depth, and interpretability over existing multi-agent counseling systems, indicating that XInsight provides a practical blueprint for integrating counseling-inspired support agents into web applications for digital well-being.
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 9297-9308
Open AccessXInsight: Integrative Stage-Consistent Psychological Counseling Support Agents for Digital Well-Being
Web-based platforms are becoming a primary channel for psychological support, yet most LLM-driven chatbots remain opaque, single-stage, and weakly grounded in established therapeutic practice. To address this gap, we present XInsight, a multi-agent framework that models psychological support as a stage-consistent workflow aligned with the classical Exploration-Insight-Action paradigm. Building on structured client representations, XInsight orchestrates specialized agents under a unified Reason-Intervene-Reflect cycle: an Exploration agent organizes background and concerns into a structured Case Conceptualization Form, a Routing agent performs Adaptive Therapeutic Routing (ATR) across SFBT, CBT, and MBCT, a unified Therapeutic agent executes school-consistent submodules, and a Consolidation agent guides review, skill integration, and relapse-prevention planning. A Recording agent continuously transforms open-ended web dialogues into standardized psychological artifacts, enhancing interpretability, continuity, and accountability. To support transparent assessment, we introduce XInsight-Bench with a Scale-Guided LLM Evaluation (SGLE) protocol that combines therapy-specific clinical scales with general counseling criteria. Experiments show improved paradigm alignment, multi-therapy integration, interaction depth, and interpretability over existing multi-agent counseling systems, indicating that XInsight provides a practical blueprint for integrating counseling-inspired support agents into web applications for digital well-being.
- 16 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (4) , 5024-5037
TG4MM: Time-Varying Gaussian Splatting for 3D Motion Magnification
3D motion magnification aims to enable us to visualize subtle, imperceptible motions by integrating eulerian video magnification with novel view synthesis. Existing method extracts the variation of feature embeddings using Neural Radiance Fields (NeRF) over time. However, this volume rendering technique suffers from two shortcomings for 3D motion magnification: (1) When reconstructing time-varying scenes through volume rendering, spatial-temporal operations between static and dynamic representations often generate noticeable artifacts, leading to blurred magnified frames. (2) When processing high-resolution dynamic scenes, the intrinsically low rendering efficiency of these techniques causes excessive computational latency, preventing real-time visualization. In this work, instead of NeRF, we propose a novel Time-varying Gaussian Splatting for 3D Motion Magnification (TG4MM) that is capable of achieving real-time rendering while effectively handling blurred magnified frames in dynamic 3D motion magnification scenes. Specifically, we propose a motion-space decoupled triplane modeling approach. The space triplane captures major spatial structures from the first frame, while the motion triplane captures subtle motion information from subsequent frames. Furthermore, we develop a phase-based motion magnification module that enhances subtle motions by applying filters within the embedding space and subtle motion triplane. Experimental results demonstrate the effectiveness of our method, showing that it outperforms existing 3D motion magnification techniques and achieves a speed up to 126 FPS.
- 16 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (4) , 5024-5037
TG4MM: Time-Varying Gaussian Splatting for 3D Motion Magnification
3D motion magnification aims to enable us to visualize subtle, imperceptible motions by integrating eulerian video magnification with novel view synthesis. Existing method extracts the variation of feature embeddings using Neural Radiance Fields (NeRF) over time. However, this volume rendering technique suffers from two shortcomings for 3D motion magnification: (1) When reconstructing time-varying scenes through volume rendering, spatial-temporal operations between static and dynamic representations often generate noticeable artifacts, leading to blurred magnified frames. (2) When processing high-resolution dynamic scenes, the intrinsically low rendering efficiency of these techniques causes excessive computational latency, preventing real-time visualization. In this work, instead of NeRF, we propose a novel Time-varying Gaussian Splatting for 3D Motion Magnification (TG4MM) that is capable of achieving real-time rendering while effectively handling blurred magnified frames in dynamic 3D motion magnification scenes. Specifically, we propose a motion-space decoupled triplane modeling approach. The space triplane captures major spatial structures from the first frame, while the motion triplane captures subtle motion information from subsequent frames. Furthermore, we develop a phase-based motion magnification module that enhances subtle motions by applying filters within the embedding space and subtle motion triplane. Experimental results demonstrate the effectiveness of our method, showing that it outperforms existing 3D motion magnification techniques and achieves a speed up to 126 FPS.
- 10 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 13 (3) , 3794-3809
Modeling Long-Term Emotional Support Through Causal World Modeling With Imitation Learning
Emotional support conversation systems have emerged as a promising complement to traditional mental health consultations, offering context-aware dialogue to support seekers’ emotional well-being. Despite their potential, two fundamental challenges remain unresolved: 1) modeling long-term emotional trajectory beyond short-term relief; and 2) adapting support strategies to context in a psychologically coherent manner. To address these challenges, we propose CAIWO, a novel framework that integrates world modeling and causality-enhanced imitation learning to systematically support seekers, alleviate psychological stress, and restore emotional balance. Specifically, CAIWO comprises two core components. The first is an emotional world model, which captures long-term emotional trajectories from historical interactions to inform anticipatory guidance. The second is a causality-enhanced imitation learning module, which infers latent causal dependencies to facilitate coherent strategy transitions and mitigate compounding errors typical of conventional imitation learning. By incorporating the final latent variables into the response decoder, CAIWO dynamically adjusts the strategies and generates emotionally resonant responses. Extensive experiments on the ESConv benchmark demonstrate that CAIWO outperforms state-of-the-art baselines by 8.6%, significantly improving the generation of responses that align with seekers’ emotional development and psychological needs.
- 10 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 13 (3) , 3794-3809
Modeling Long-Term Emotional Support Through Causal World Modeling With Imitation Learning
Emotional support conversation systems have emerged as a promising complement to traditional mental health consultations, offering context-aware dialogue to support seekers’ emotional well-being. Despite their potential, two fundamental challenges remain unresolved: 1) modeling long-term emotional trajectory beyond short-term relief; and 2) adapting support strategies to context in a psychologically coherent manner. To address these challenges, we propose CAIWO, a novel framework that integrates world modeling and causality-enhanced imitation learning to systematically support seekers, alleviate psychological stress, and restore emotional balance. Specifically, CAIWO comprises two core components. The first is an emotional world model, which captures long-term emotional trajectories from historical interactions to inform anticipatory guidance. The second is a causality-enhanced imitation learning module, which infers latent causal dependencies to facilitate coherent strategy transitions and mitigate compounding errors typical of conventional imitation learning. By incorporating the final latent variables into the response decoder, CAIWO dynamically adjusts the strategies and generates emotionally resonant responses. Extensive experiments on the ESConv benchmark demonstrate that CAIWO outperforms state-of-the-art baselines by 8.6%, significantly improving the generation of responses that align with seekers’ emotional development and psychological needs.
- Top Cited Papers
- 3 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 4624-4637
Task-Generalized Adaptive Cross-Domain Learning for Multimodal Image Fusion
Multimodal Image Fusion (MMIF) aims to integrate complementary information from different imaging modalities to overcome the limitations of individual sensors. It enhances image quality and facilitates downstream applications such as remote sensing, medical diagnostics, and robotics. Despite significant advancements, current MMIF methods still face challenges such as modality misalignment, high-frequency detail destruction, and task-specific limitations. To address these challenges, we propose AdaSFFuse, a novel framework for task-generalized MMIF through adaptive cross-domain co-fusion learning. AdaSFFuse introduces two key innovations: the Adaptive Approximate Wavelet Transform (AdaWAT) for frequency decoupling, and the Spatial-Frequency Mamba Blocks for efficient multimodal fusion. AdaWAT adaptively separates the high- and low-frequency components of multimodal images from different scenes, enabling fine-grained extraction and alignment of distinct frequency characteristics for each modality. The Spatial-Frequency Mamba Blocks facilitate cross-domain fusion in both spatial and frequency domains, enhancing this process. These blocks dynamically adjust through learnable mappings to ensure robust fusion across diverse modalities. By combining these components, AdaSFFuse improves the alignment and integration of multimodal features, reduces frequency loss, and preserves critical details. Extensive experiments on four MMIF tasks-Infrared-Visible Image Fusion (IVF), Multi-Focus Image Fusion (MFF), Multi-Exposure Image Fusion (MEF), and Medical Image Fusion (MIF)-demonstrate AdaSFFuse's superior fusion performance, ensuring both low computational cost and a compact network, offering a strong balance between performance and efficiency. The code will be publicly available at https://github.com/Zhen-yu-Liu/AdaSFFuse.
- 3 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 4624-4637
Top Cited PapersTask-Generalized Adaptive Cross-Domain Learning for Multimodal Image Fusion
Multimodal Image Fusion (MMIF) aims to integrate complementary information from different imaging modalities to overcome the limitations of individual sensors. It enhances image quality and facilitates downstream applications such as remote sensing, medical diagnostics, and robotics. Despite significant advancements, current MMIF methods still face challenges such as modality misalignment, high-frequency detail destruction, and task-specific limitations. To address these challenges, we propose AdaSFFuse, a novel framework for task-generalized MMIF through adaptive cross-domain co-fusion learning. AdaSFFuse introduces two key innovations: the Adaptive Approximate Wavelet Transform (AdaWAT) for frequency decoupling, and the Spatial-Frequency Mamba Blocks for efficient multimodal fusion. AdaWAT adaptively separates the high- and low-frequency components of multimodal images from different scenes, enabling fine-grained extraction and alignment of distinct frequency characteristics for each modality. The Spatial-Frequency Mamba Blocks facilitate cross-domain fusion in both spatial and frequency domains, enhancing this process. These blocks dynamically adjust through learnable mappings to ensure robust fusion across diverse modalities. By combining these components, AdaSFFuse improves the alignment and integration of multimodal features, reduces frequency loss, and preserves critical details. Extensive experiments on four MMIF tasks-Infrared-Visible Image Fusion (IVF), Multi-Focus Image Fusion (MFF), Multi-Exposure Image Fusion (MEF), and Medical Image Fusion (MIF)-demonstrate AdaSFFuse's superior fusion performance, ensuring both low computational cost and a compact network, offering a strong balance between performance and efficiency. The code will be publicly available at https://github.com/Zhen-yu-Liu/AdaSFFuse.
- 10 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-11
Defocused Camera Calibration With Self-Encoded Phase-Shifting Circular Grating Arrays
Camera calibration plays an important role in machine vision systems as it directly impacts the measurement accuracy. However, traditional calibration methods usually perform poorly when calibrating defocused cameras. To overcome this issue, this paper proposes an effective defocused camera calibration method that enables feature detection with high accuracy from blurred target images. This study designs a new calibration target called as self-encoded phase-shifting circular grating (PCG) arrays. Phase shifts of each PCG are used to encode its index, and a robust decoding algorithm is developed to identify each PCG. The phase map is recovered via phase-shifting algorithms, and the PCG centers are extracted using minimum projection and least-square fitting algorithms. Based on these PCG centers and their indices, the projection mapping between the 3D world and 2D image can be established for accurate estimation of internal parameters. Compared with traditional methods, our method utilizes more valuable phase points to precisely locate the PCG centers, and utilizes phase shifts to correctly identify the PCG indices. Simulations and experiments have been conducted to validate the proposed method, demonstrating its superior performance over traditional calibration methods.
- 10 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-11
Defocused Camera Calibration With Self-Encoded Phase-Shifting Circular Grating Arrays
Camera calibration plays an important role in machine vision systems as it directly impacts the measurement accuracy. However, traditional calibration methods usually perform poorly when calibrating defocused cameras. To overcome this issue, this paper proposes an effective defocused camera calibration method that enables feature detection with high accuracy from blurred target images. This study designs a new calibration target called as self-encoded phase-shifting circular grating (PCG) arrays. Phase shifts of each PCG are used to encode its index, and a robust decoding algorithm is developed to identify each PCG. The phase map is recovered via phase-shifting algorithms, and the PCG centers are extracted using minimum projection and least-square fitting algorithms. Based on these PCG centers and their indices, the projection mapping between the 3D world and 2D image can be established for accurate estimation of internal parameters. Compared with traditional methods, our method utilizes more valuable phase points to precisely locate the PCG centers, and utilizes phase shifts to correctly identify the PCG indices. Simulations and experiments have been conducted to validate the proposed method, demonstrating its superior performance over traditional calibration methods.
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 5461-5470
Motion Matters: Motion-guided Modulation Network for Skeleton-based Micro-Action Recognition
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 5461-5470
Open AccessMotion Matters: Motion-guided Modulation Network for Skeleton-based Micro-Action Recognition
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 14216-14221
MAC 2025: The 2nd Micro-Action Analysis Grand Challenge
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 14216-14221
Open AccessMAC 2025: The 2nd Micro-Action Analysis Grand Challenge