Fuchun Sun
YOU?
Author Swipe
View article: Tongji Biobank: A High-Quality Repository Integrating Multi-Dimensional Biomedical Data
Tongji Biobank: A High-Quality Repository Integrating Multi-Dimensional Biomedical Data Open
With the advancement of biotechnology and computer technology, life sciences and medical data have grown rapidly, and research paradigms have undergone revolutionary changes. Biomedical and clinical research have entered the big data era, …
View article: Audio-Guided Visual Perception for Audio-Visual Navigation
Audio-Guided Visual Perception for Audio-Visual Navigation Open
Audio-Visual Embodied Navigation aims to enable agents to autonomously navigate to sound sources in unknown 3D environments using auditory cues. While current AVN methods excel on in-distribution sound sources, they exhibit poor cross-sour…
View article: ECTSpeech: Enhancing Efficient Speech Synthesis via Easy Consistency Tuning
ECTSpeech: Enhancing Efficient Speech Synthesis via Easy Consistency Tuning Open
Diffusion models have demonstrated remarkable performance in speech synthesis, but typically require multi-step sampling, resulting in low inference efficiency. Recent studies address this issue by distilling diffusion models into consiste…
View article: EGSTalker: Real-Time Audio-Driven Talking Head Generation with Efficient Gaussian Deformation
EGSTalker: Real-Time Audio-Driven Talking Head Generation with Efficient Gaussian Deformation Open
This paper presents EGSTalker, a real-time audio-driven talking head generation framework based on 3D Gaussian Splatting (3DGS). Designed to enhance both speed and visual fidelity, EGSTalker requires only 3-5 minutes of training video to s…
View article: Audio-Guided Dynamic Modality Fusion with Stereo-Aware Attention for Audio-Visual Navigation
Audio-Guided Dynamic Modality Fusion with Stereo-Aware Attention for Audio-Visual Navigation Open
In audio-visual navigation (AVN) tasks, an embodied agent must autonomously localize a sound source in unknown and complex 3D environments based on audio-visual signals. Existing methods often rely on static modality fusion strategies and …
View article: Advancing Audio-Visual Navigation Through Multi-Agent Collaboration in 3D Environments
Advancing Audio-Visual Navigation Through Multi-Agent Collaboration in 3D Environments Open
Intelligent agents often require collaborative strategies to achieve complex tasks beyond individual capabilities in real-world scenarios. While existing audio-visual navigation (AVN) research mainly focuses on single-agent systems, their …
View article: PGSTalker: Real-Time Audio-Driven Talking Head Generation via 3D Gaussian Splatting with Pixel-Aware Density Control
PGSTalker: Real-Time Audio-Driven Talking Head Generation via 3D Gaussian Splatting with Pixel-Aware Density Control Open
Audio-driven talking head generation is crucial for applications in virtual reality, digital avatars, and film production. While NeRF-based methods enable high-fidelity reconstruction, they suffer from low rendering efficiency and suboptim…
View article: Study on Loss of Necessary Benefit in Administrative Compensation
Study on Loss of Necessary Benefit in Administrative Compensation Open
On March 21, 2022, the Supreme People’s Court issued the Judicial Interpretation on Administrative Compensation, expanding the scope of direct losses under the State Compensation Law. This landmark development extends the definition of act…
View article: Global research trends on artificial intelligence in psychological interventions for stroke survivors: a bibliometric and visualized analysis (2000–2024)
Global research trends on artificial intelligence in psychological interventions for stroke survivors: a bibliometric and visualized analysis (2000–2024) Open
Objective This study aimed to conduct a bibliometric analysis of research literature on AI-assisted psychological interventions for stroke survivors published from 2000 to 2024, using CiteSpace and VOSviewer to examine research collaborati…
View article: Flow-Based Policy for Online Reinforcement Learning
Flow-Based Policy for Online Reinforcement Learning Open
We present \textbf{FlowRL}, a novel framework for online reinforcement learning that integrates flow-based policy representation with Wasserstein-2-regularized optimization. We argue that in addition to training signals, enhancing the expr…
View article: On the Transferability and Discriminability of Repersentation Learning in Unsupervised Domain Adaptation
On the Transferability and Discriminability of Repersentation Learning in Unsupervised Domain Adaptation Open
In this paper, we addressed the limitation of relying solely on distribution alignment and source-domain empirical risk minimization in Unsupervised Domain Adaptation (UDA). Our information-theoretic analysis showed that this standard adve…
View article: A phase search-enhanced Bi-RRT path planning algorithm for mobile robots
A phase search-enhanced Bi-RRT path planning algorithm for mobile robots Open
The proposed improvement to the Rapidly-exploring Random Tree (RRT) path planning algorithm is aimed at addressing the issue of slow convergence speed caused by boundary information in the original algorithm, by introducing a phase search …
View article: Self-Supervised Enhancement of Forward-Looking Sonar Images: Bridging Cross-Modal Degradation Gaps through Feature Space Transformation and Multi-Frame Fusion
Self-Supervised Enhancement of Forward-Looking Sonar Images: Bridging Cross-Modal Degradation Gaps through Feature Space Transformation and Multi-Frame Fusion Open
Enhancing forward-looking sonar images is critical for accurate underwater target detection. Current deep learning methods mainly rely on supervised training with simulated data, but the difficulty in obtaining high-quality real-world pair…
View article: Evolutionary Reinforcement Learning with Parameterized Action Primitives for Diverse Manipulation Tasks
Evolutionary Reinforcement Learning with Parameterized Action Primitives for Diverse Manipulation Tasks Open
Reinforcement learning (RL) has shown promising performance in tackling robotic manipulation tasks (RMTs), which require learning a prolonged sequence of manipulation actions to control robots efficiently. However, most RL algorithms often…
View article: Siamese Foundation Models for Crystal Structure Prediction
Siamese Foundation Models for Crystal Structure Prediction Open
Crystal Structure Prediction (CSP), which aims to generate stable crystal structures from compositions, represents a critical pathway for discovering novel materials. While structure prediction tasks in other domains, such as proteins, hav…
View article: Comprehensive analysis of anoikis-related gene signature in ulcerative colitis using machine learning algorithms
Comprehensive analysis of anoikis-related gene signature in ulcerative colitis using machine learning algorithms Open
Ulcerative colitis (UC) is a chronic inflammatory bowel disease with an idiopathic origin, characterized by persistent mucosal inflammation. Anoikis is a programmed cell death mechanism activated during carcinogenesis to eliminate undetect…
View article: Time Series Domain Adaptation via Latent Invariant Causal Mechanism
Time Series Domain Adaptation via Latent Invariant Causal Mechanism Open
Time series domain adaptation aims to transfer the complex temporal dependence from the labeled source domain to the unlabeled target domain. Recent advances leverage the stable causal mechanism over observed variables to model the domain-…
View article: One-Off Irrigation Combined Subsoiling and Nitrogen Management Enhances Wheat Grain Yield by Optimizing Physiological Characteristics in Leaves in Dryland Regions
One-Off Irrigation Combined Subsoiling and Nitrogen Management Enhances Wheat Grain Yield by Optimizing Physiological Characteristics in Leaves in Dryland Regions Open
Irrigation practice, tillage method, and nitrogen (N) management are the three most important agronomic measures for wheat (Triticum aestivum L.) production, but the combined effects on grain yield and wheat physiological characteristics a…
View article: Medical Multimodal Foundation Models in Clinical Diagnosis and Treatment: Applications, Challenges, and Future Directions
Medical Multimodal Foundation Models in Clinical Diagnosis and Treatment: Applications, Challenges, and Future Directions Open
Recent advancements in deep learning have significantly revolutionized the field of clinical diagnosis and treatment, offering novel approaches to improve diagnostic precision and treatment efficacy across diverse clinical domains, thus dr…
View article: A Comprehensive Survey on Embodied Intelligence: Advancements, Challenges, and Future Perspectives
A Comprehensive Survey on Embodied Intelligence: Advancements, Challenges, and Future Perspectives Open
Embodied Intelligence, which integrates physical interaction capabilities with cognitive computation in real-world scenarios, provides a promising path to achieve Artificial General Intelligence (AGI). Recently, the landscape of embodied i…
View article: Artificial Skin Based on Visuo‐Tactile Sensing for 3D Shape Reconstruction: Material, Method, and Evaluation
Artificial Skin Based on Visuo‐Tactile Sensing for 3D Shape Reconstruction: Material, Method, and Evaluation Open
Artificial skin has shown great potential in robot perception and human healthcare. It provides multifunctional tactile sensing, including 3D shape reconstruction, contact feedback, and temperature perception, where the 3D reconstruction f…
View article: On the Generalization and Causal Explanation in Self-Supervised Learning
On the Generalization and Causal Explanation in Self-Supervised Learning Open
Self-supervised learning (SSL) methods learn from unlabeled data and achieve high generalization performance on downstream tasks. However, they may also suffer from overfitting to their training data and lose the ability to adapt to new ta…