Age significantly influences human motor patterns, yet existing virtual reality (VR) systems lack dynamic modelling of these variations. This paper introduces AgeStyle, a versatile framework that integrates age-guided style selection with motion style transfer to convert user-uploaded videos into interactive 3D motion models. Utilizing 2D joint detection and 3D pose estimation, AgeStyle constructs motion representations enhanced by a CLIP-driven Cross-Attention module, capturing the distinct traits of different age groups—child flexibility, adult efficiency, and elderly stability. Our system enables real-time switching between motion styles and perspectives through voice commands, offering an immersive exploration of age-related movements. Quantitative experiments on the XIA dataset demonstrate AgeStyle’s competitive performance in both content preservation and style consistency, achieving average CC and SC++ scores of 7.4 and 14.8, respectively. AgeStyle represents a meaningful advancement in VR character design, with broad potential applications in education, healthcare, rehabilitation, and interactive entertainment. For the demo, please refer to https://youtu.be/eo7Shy0Ukps. The source code of AgeStyle is available at https://github.com/codeozzz/ageStyle.
Large-area semi-closed holes on the lateral surfaces of tooth models can cause difficulties in subsequent orthodontic tasks, such as collision detection and finite element analysis, and may affect the accuracy and stability of further processing. To address this issue, this paper proposes a three-dimensional tooth semi-closed hole repair method based on the moving least squares technique. First, key points were manually selected to transform semi-closed holes into closed ones, and new vertices were uniformly distributed within the hole region to construct an initial triangular mesh. Subsequently, a mesh homogenization algorithm was applied to improve the quality of the generated triangular patches. Then, a moving least-squares surface-fitting approach was employed. By incorporating the point cloud density and Gaussian curvature, an adaptive vertex influence region was designed to iteratively adjust the vertex positions and better fit the target surface. Finally, the generated mesh was smoothed using the Laplacian-Beltrami operator to obtain a repaired surface consistent with the original geometry. The experimental results demonstrate that the proposed method achieves superior efficiency and repair quality compared to existing approaches.
Cognitive decline (CD) is a hallmark symptom of many neurodegenerative diseases, and in severe cases, it progresses to cognitive dysfunction and a loss of independent daily functioning. Eye movement interaction (EMI) has shown promise in detecting early signs of CD, but its applications remain largely confined to clinical settings, limiting integration into everyday life. In response, this paper proposes a novel method based on naturalistic EMI for the prediagnosis of CD, leveraging sports streaming to induce saccades and using ball movements as dynamic stimuli to uncover early neural dysfunction. Through a pilot and empirical study, we demonstrate that ball games can effectively elicit saccades, and our algorithm can accurately extract batting saccades to distinguish between the healthy control group (HCG) and the cognitive decline group (CDG). In this study with 60 participants, the accuracy of recognizing CDG in natural, unmarked game viewing was 0.7333 at best, and reached 0.8833 when semi-natural random flashing cues were introduced on balls. These findings illustrate the potential for seamlessly integrating naturalistic EMI-based prediagnosis into everyday interactions.
Craniofacial reconstruction aims to estimate a person’s facial appearance from the skull and has applications in forensic investigation, reconstructive surgery, archaeology, and anthropology. The skull and face are typically represented as triangular meshes, on which geodesics provide intrinsic surface structures. Accordingly, we propose a craniofacial reconstruction framework that represents facial geometry using geodesics and learns a geodesic-based statistical model. First, we extract a set of geodesic curves for each registered face mesh, originating from the nose tip. Next, paired skull meshes and their corresponding facial geodesics are used to train a principal component analysis (PCA)-based statistical model that predicts the facial geodesics for an unseen skull. Finally, we reconstruct the full-face mesh by fitting a facial statistical shape model to the predicted geodesics. Because the intrinsic properties of the geodesics and geodesic distances are preserved under isometric deformations, the accuracy of the corresponding feature points automatically established by the geodesics and exact correspondence between the feature points are ensured, thereby improving the accuracy of the craniofacial reconstruction results. Experiments demonstrate improved reconstruction accuracy and reduced runtime compared with baseline methods.
This survey reviews feed-forward, point-cloud 3D reconstruction methods from DUSt3R to VGGT and their recent variants. Here, feed-forward primarily refers to predicting dense geometry and, when applicable, camera poses through learned network inference, without relying on classical per-scene SfM + MVS optimization as the main inference mechanism. We first formalize the reconstruction task in pose-aware and pose-free settings, and contrast feed-forward point-map regression with classical Structure-from-Motion and Multi-View Stereo pipelines. Building on this, we organize existing methods into three stages: early pairwise models typified by DUSt3R, DUSt3R-style extensions that enhance multi-view consistency, streaming, efficiency, and dynamic-scene handling, and large unified transformers such as VGGT that process tens to hundreds of views jointly, while noting differences in their inference paradigms. We analyze these models along shared axes, including scene representation, correspondence reasoning, pose regression, fusion strategies, and the role of large-scale training data. We summarize widely used 3D datasets and evaluation metrics, and provide a case study on the DTU benchmark for multi-view depth and point map estimation, highlighting accuracy-efficiency trade-offs between optimization-based and feed-forward approaches. Finally, we discuss open challenges in data scarcity, sparse-view reconstruction, non-Lambertian structures, dynamic scenes, long-context processing, and resource-efficient deployment, and outline future directions that combine feed-forward architectures with differentiable rendering, generative priors, and safety mechanisms to enable scalable and trustworthy 3D reconstruction systems.
Solving partial differential equations (PDEs) for fluid simulation is computationally expensive, especially when dealing with complex geometries and high-resolution meshes. Recent advances in physics-informed graph neural networks (PIGNNs) have demonstrated potential in approximating such simulations more efficiently. Particularly, thermodynamic informed graph neural networks (TIGNNs) offer a promising data-driven alternative to traditional PDE solvers for fluid simulations. However, existing TIGNN implementations suffer from significant training inefficiencies, requiring prolonged runtimes and high memory consumption due to the need to maintain large parameter matrices in GPU memory. Inspired by multi-processor strategies in deformable solid simulations, we propose a novel multi-processor thermodynamic-informed graph neural network (MP-TIGNN) architecture to significantly accelerate training without compromising accuracy. Our approach enables faster convergence and reduces memory usage with mixed-precision training by leveraging fully sharded data parallelism (FSDP) across multiple GPUs. Experimental results show that our approach reduces training time by approximately 70% compared to the original setup while maintaining similar prediction accuracy.
Reconstructing accurate urban building models remains challenging because of large-scale variability and complex topologies. We introduce a practical multi-step framework that reconstructs a single building from a dense triangular mesh by explicitly exploiting its block-wise composition. The pipeline first partitioned the mesh into spatially coherent regions, then extracted height-aware contours, and classified the geometry into facades, roofs, and appurtenances. For each part, we performed contour-guided vectorized modeling with profile fitting and consistency constraints, producing watertight, semantically structured models at the level of detail (LOD) 2.3. The interactive refinement module further supported user adjustments to resolve rare failures and enforced design regularity. The proposed decomposition and layered fitting yielded compact outputs while preserving the salient geometry and made the reconstruction robust across scales and moderate noise. We also analyzed the modeling assumptions and implementation details to ensure reproducibility. Overall, this study offers a component-aware, engineering-ready solution for converting unstructured meshes into structured building models that are amenable to downstream urban modeling and visualization.
Accurately representing and rendering dynamic scenes over time remains a central challenge in neural rendering and computer graphics. Existing dynamic Gaussian-based methods often suffer from limited temporal consistency, flickering under fast motion, and poor adaptability to non-human structures. To address these issues, we propose DG-4DGS, a deformation-graph-constrained 4D Gaussian splatting framework for temporally stable dynamic rendering. The method anchors all Gaussians in a canonical space and enforces cross-frame geometric alignment through a deformation graph. Based on neighborhood-consistency features, a multi-head residual decoder refines position, rotation/scale, and color attributes to achieve fine-detail fidelity without relying on online densification or pruning. Compared with 4DGS and avatar-based approaches, DG-4DGS achieves higher PSNR (peak signal-to-noise ratio) and SSIM (structural similarity index measure) scores and significantly smaller model size on both the TalkBody4D (human) and Horse (non-human) datasets. It effectively suppresses temporal flickering and cross-frame drift in high-frequency regions such as hair strands, cloth wrinkles, and limb extremities. The framework does not depend on parametric templates, facilitating extension to non-human and complex clothing scenarios, though its performance still depends on deformation-tracking quality and neighborhood topology selection.
3D Gaussian splatting (3DGS) has gained significant attention for its real-time, photorealistic rendering in novel view synthesis. However, its performance degrades severely when applied to real-world scenes with transients that break cross-view consistency. Existing methods typically attempt to identify and mask out these transients, but inherent masking errors often leave behind incoherent floating Gaussians, resulting in spotted artifacts that degrade the final scene quality. To address these limitations, we propose CleanSplat, a curriculum structural Gaussian splatting method that purifies the scene Gaussians through curriculum 3DGS optimization for transient removal and structural pruning of spotted artifacts. We introduce a curriculum-guided masking paradigm that generates coarse-to-fine transient masks from multi-scale features. The progressive optimization is driven by modulating the masking supervision based on current training state. To clear the spots, we propose a structure-aware handling strategy that employs a superpoint graph (SPG) partitioning of the Gaussians to perform principled identification and hierarchical pruning. This allows for the filtering of both intra-superpoint outliers and entire spurious superpoints based on local 3D coherence instead of only simple photometric consistency. By integrating curriculum 3DGS optimization and structural pruning, our method effectively separates the transients and purifies the static scene Gaussians. Extensive experiments on challenging datasets demonstrate that CleanSplat significantly outperforms state-of-the-art methods, delivering more detailed and cleaner novel view synthesis.
We present a novel 360° panoramic video conferencing system that dynamically synchronizes virtual backgrounds with real-time camera motion, addressing the limitations of static backgrounds in conventional systems. By integrating robust human segmentation, monocular visual odometry (VO), and virtual environment rendering, our method achieves seamless alignment between foreground participants and immersive 3D virtual scenes. Unlike prior approaches that suffer from foreground-background desynchronization during camera rotations or user movements, our framework estimates camera rotation in 3-DoF using a hybrid pipeline combining feature-based patch tracking and pose smoothing, while ignoring translation artifacts to maintain stability. This work bridges the gap between computational efficiency and MR-driven telepresence, offering a practical solution for next-generation virtual collaboration.
Surgical navigation has evolved significantly through advances in augmented reality, virtual reality, and mixed reality, improving precision and safety across many clinical applications, including neurosurgery, maxillofacial, spinal, and arthroplasty procedures. By integrating preoperative imaging with real-time intraoperative data, these systems provide dynamic guidance, reduce radiation exposure, and minimize tissue damage. Key challenges persist, including intraoperative registration accuracy, flexible tissue deformation, respiratory compensation, and real-time imaging quality. Emerging solutions include artificial intelligence-driven segmentation, deformation-field modeling, and hybrid registration techniques. Future developments will include lightweight, portable systems, improved non-rigid registration algorithms, and greater clinical adoption. Despite advances in rigid-tissue applications, soft-tissue navigation requires additional innovation to address motion variability and registration reliability, ultimately advancing minimally invasive surgery and precision medicine.
Digital twin technology, that creates virtual replicas of physical entities using real-time data and simulation models, has emerged as a transformative innovation across multiple healthcare domains. Its application in physiotherapy and rehabilitation represents a paradigm shift from traditional therapeutic approaches to personalized data-driven interventions that optimize patient outcomes. This narrative review examines the current applications, benefits, challenges, and future prospects of digital twin technology in physiotherapy and rehabilitation, providing a comprehensive analysis of the manner in which this technology is reshaping clinical practice and patient care. A narrative review approach was employed, systematically searching PubMed, IEEE Xplore, Scopus, and Web of Science databases. Studies describing digital twin applications, development methodologies, clinical implementations, and theoretical frameworks in physiotherapy and rehabilitation contexts were included. Digital twin technology demonstrates significant potential in personalizing rehabilitation programs, enabling real-time monitoring of patient progress, predicting treatment outcomes, and facilitating remote therapeutic interventions. Current applications span musculoskeletal rehabilitation, neurological recovery, post-surgical care, and sports injury management. Key benefits include enhanced treatment precision, improved patient engagement, reduced healthcare costs, and accelerated recovery times. However, implementation faces challenges including technological complexity, data privacy concerns, interoperability issues, and the need for substantial infrastructure investment. Digital twin technology represents a promising frontier in physiotherapy and rehabilitation, offering unprecedented opportunities for personalized, efficient, and effective patient care. Successful integration requires addressing the current limitations while fostering interdisciplinary collaboration between clinicians, engineers, and data scientists.
Background: Computed tomography (CT) and cone-beam computed tomography (CBCT) image registration play pivotal roles in computer-assisted navigation for orthopedic surgery. Traditional methods often apply uniform deformation models, neglecting the biomechanical differences between rigid structures and soft tissues, which compromises registration accuracy, especially during significant bone displacements. Method: To address this issue, we introduce RE-Reg, a rigid-elastic CT-CBCT image registration framework that jointly learns rigid bone motion and soft tissue deformation. RE-Reg incorporates a rigid alignment (RA) module to estimate global bone motion and an elastic deformation (ED) module to model soft tissue deformation, preserving bony structures through bone shape preservation (BSP) loss. Result: Our comprehensive evaluation on publicly available datasets demonstrates that RE-Reg significantly outperforms existing methods in terms of registration accuracy and rigid bone structure preservation, achieving a 1.3% improvement in Dice similarity coefficient (DSC) and a 23% reduction in rigid bone deformation (%Δvol) compared with the best baseline. Conclusion: This framework not only enhances anatomical fidelity but also ensures biomechanical plausibility and provides a valuable tool for image-guided orthopedic surgery. This code is available at https://github.com/Zq-Huang/RE-Reg.
Background: 3D botanical tree reconstruction from a single image plays a vital role in the field of computer graphics. However, accurately capturing the intricate branching patterns and detailed morphologies of trees remains a challenge. Methods: In this study, we proposed a novel approach for single-image tree reconstruction using a conditional generative adversarial network to infer the 3D skeleton of a tree in the form of a 2D skeleton depth map. Based on the 2D skeleton depth map, a corresponding branching structure (3D skeleton) that inherits the tree shape in the input image and leaves can be generated using a procedural modeling technique. Result: Experimental results show that the proposed method accurately reconstructs diverse tree structures across species. Both quantitative and qualitative evaluations demonstrate improved skeleton completeness, branching accuracy, and visual realism over baseline methods, while requiring no user input. Conclusions: Our proposed approach for generating lifelike 3D tree models from a single image with no user input shows its proficiency in achieving efficient and reliable reconstruction. These results showcase the capability of the proposed model to recreate complex tree architectures while capturing their visual authenticity.
Optical coherence tomography (OCT), particularly Swept-Source OCT, is widely employed in medical diagnostics and industrial inspections owing to its high-resolution imaging capabilities. However, Swept-Source OCT 3D imaging often suffers from stripe artifacts caused by unstable light sources, system noise, and environmental interference, posing challenges to real-time processing of large-scale datasets. To address this issue, this study introduces a real-time reconstruction system that integrates stripe-artifact suppression and parallel computing using a graphics processing unit. This approach employs a frequency-domain filtering algorithm with adaptive anti-suppression parameters, dynamically adjusted through an image quality evaluation function and optimized using a convolutional neural network for complex frequency-domain feature learning. Additionally, a graphics processing unit integrated 3D reconstruction framework is developed, enhancing data processing throughput and real-time performance via a dual-queue decoupling mechanism. Experimental results demonstrate significant improvements in structural similarity (0.92), peak signal-to-noise ratio (31.62 dB), and stripe suppression ratio (15.73 dB) compared with existing methods. On the RTX 4090 platform, the proposed system achieved an end-to-end delay of 94.36 milliseconds, a frame rate of 10.3 frames per second, and a throughput of 121.5 million voxels per second, effectively suppressing artifacts while preserving image details and enhancing real-time 3D reconstruction performance.
LiDAR and camera are two of the most common sensors used in the fields of robot perception, autonomous driving, augmented reality, and virtual reality, where these sensors are widely used to perform various tasks such as odometry estimation and 3D reconstruction. Fusing the information from these two sensors can significantly increase the robustness and accuracy of these perception tasks. The extrinsic calibration between cameras and LiDAR is a fundamental prerequisite for multimodal systems. Recently, extensive studies have been conducted on the calibration of extrinsic parameters. Although several calibration methods facilitate sensor fusion, a comprehensive summary for researchers and, especially, non-expert users is lacking. Thus, we present an overview of extrinsic calibration and discuss diverse calibration methods from the perspective of calibration system design. Based on the calibration information sources, this study classifies these methods as target-based or targetless. For each type of calibration method, further classification was performed according to the diverse types of features or constraints used in the calibration process, and their detailed implementations and key characteristics were introduced. Thereafter, calibration-accuracy evaluation methods are presented. Finally, we comprehensively compare the advantages and disadvantages of each calibration method and suggest directions for practical applications and future research.
Intrinsic decomposition, the process of decomposing an image into reflectance and shading, is widely used in virtual and augmented reality tasks. Reflectance and shading often exhibit large gradients at the object edges, and the intrinsic properties on the same object tend to be similar. This spatial coherence is closely related to semantic consistency because objects within the same semantic category often exhibit similar intrinsic properties. Therefore, incorporating semantic segmentation into a deep intrinsic decomposition framework helps the network distinguish between different object instances and understand high-level scene structures. To this end, we design an intrinsic decomposition network jointly trained with a dedicated semantic segmentation module, allowing semantic cues to enhance the decomposition of reflectance and shading. The semantic module provides guidance during training but is removed during inference, improving performance without increasing the inference cost. Additionally, to capture the global contextual dependencies critical for intrinsic decomposition, we adopt a Transformer-based backbone. The proposed backbone enables the model to associate distant regions with similar material properties, thereby maintaining consistency in reflectance and learning smooth illumination patterns across a scene. A convolutional decoder is also designed to output predictions with improved details. Experiments demonstrate that our approach achieves state-of-the-art performance in the quantitative evaluations on the Intrinsic Images in the Wild (IIW) and Shading Annotations in the wild (SAW) datasets.
Compared with traditional open surgery, laparoscopic surgery significantly reduces bodily trauma, postoperative pain, and hospitalization duration. However, owing to the small size of incisions and the counterintuitive motion of surgical tools, longer training cycles are required for surgeons to achieve fine operational skills. This paper presents a laparoscopic surgery simulator with haptic-feedback control (LSHC-6) that provides a reliable and cost-effective training alternative for surgeons. In addition to the structural diagram, kinematic analysis, and gravity compensation algorithm, a particle swarm optimization algorithm (PSO) is applied to optimize the structural parameters of the simulator by evaluating its workspace, global dexterity, and gravity compensation ability. A prototype system was developed and evaluated using two training experiments. The results demonstrate that the simulator exhibits good operational fluidity, workspace, and stable force output, effectively meeting the needs of laparoscopic surgical training.
The integration of Human-Robot Collaboration(HRC) into Virtual Reality(VR) technology is transforming industries by enhancing workforce skills, improving safety, and optimizing operational processes and efficiency through realistic simulations of industry-specific scenarios. Despite the growing adoption of VR integrated with HRC, comprehensive reviews of current research in HRC-VR within the construction and manufacturing fields are lacking. This review examines the latest advances in designing and implementing HRC using VR technology in these industries. The aim is to address the application domains of HRC-VR, types of robots used, VR setups, and software solutions used. To achieve this, a systematic literature review using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses methodology was conducted on the Web of Science and Google Scholar databases, analyzing 383 articles and selecting 53 papers that met the established selection criteria. The findings emphasize a significant focus on enhancing human-robot interaction with a trend toward using immersive VR experiences and interactive 3D content creation tools. However, the integration of HRC with VR, especially in the dynamic construction environment, presents unique challenges and opportunities for future research, including developing more realistic simulations and adaptable robot systems. This paper offers insights for researchers, practitioners, educators, industry professionals, and policymakers interested in leveraging the integration of HRC with VR in construction and manufacturing industries.
