Accurate segmentation of cerebral aneurysms from 3D vessel meshes is an essential yet challenging task, complicated by diverse imaging modalities that produce multi-resolution representations. Existing methods often struggle to handle meshes of varying granularity and orientations while maintaining segmentation accuracy. In this paper, we propose an end-to-end multi-scale graph-based segmentation framework that incorporates SE(3)-equivariance and an uncertainty-aware loss function. Our approach constructs a multi-scale graph representation on the 3D mesh and leverages a self-attention mechanism over graph edges to achieve adaptive neighborhood awareness, enabling the network to effectively handle multiple mesh resolutions simultaneously. By introducing an SE(3)-equivariant backbone, rotational variations in aneurysm orientation are naturally accommodated, ensuring relevant and effective feature learning. Furthermore, we develop an uncertainty-aware loss that adaptively emphasizes ambiguous regions, improving segmentation quality and confidence. Experimental results on the public datasets IntrA demonstrate that our method outperforms existing techniques, offering improved accuracy, consistency and stability across different mesh resolutions for 3D aneurysm segmentation. Code is available on https://github.com/Dolphin4mi/se3meshseg.

3D human motion prediction is a challenging task due to the highly non-linear nature of movements. Existing deep learning-based action prediction methods emphasize the design of sophisticated network architectures to achieve state-of-the-art performance on datasets. However, in real-life scenarios, changes in skeletal size lead to a shift in data distribution, which presents challenges for accurate motion prediction. Additionally, the presence of stretching artifacts in predicted bone sequences significantly impacts the quality of the predictions. To address these issues, we propose a framework that combines geometric encoding with neural networks to achieve size-invariant and high-quality motion prediction without stretching artifacts. We consider the constraint of bone length and construct a motion space using Riemannian manifold theory, which remains unaffected by changes in skeletal size and can fully represent human motion. Furthermore, we propose the trajectory transport square-root velocity function to encode motion sequences into a flattened space. This transformation simplifies the distance calculation and linearizes the optimization problem in non-flattened space. Experiments on the Human 3.6M and CMU MoCap datasets demonstrated that the proposed method has achieved competitive performance without any stretching artifacts and exhibits robustness to changes in skeletal size.

Automatic detecting intracranial aneurysms (IAs) poses significant challenges due to their diversity, varying locations, and complex classifications by size, shape, and phenotype. Current shape-based IAs detection methods, while promising, often neglect the topological connectivity of IAs vertices and the variable traits of the aneurysm's neck, leading to fragmented detections. To address these issues, we present a mesh convolutional neural network based on gauge equivariant convolution to leverage the topological and geometric features of 3D mesh models. Our network comprises four key components: anisotropic message passing (AMP) on mesh surfaces, gauge equivariant convolution (GEC), vector-aware feature reconstruction (VFR), and a pooling-free convolutional architecture. AMP ensures accurate detection of IAs from surrounding vessels by utilizing topological connectivity and anisotropic relationships between mesh vertices. GEC offers rotational equivariance for consistently learning geometric features, improving feature learning stability and efficiency. VFR preserves the geometric and directional integrity of the vector features, enriching the representational capacity of the network. The pooling-free convolutional architecture captures local and global geometric nuances of 3D meshes, achieving precise IAs detection and producing sharper IAs boundaries. Tests on the IntrA dataset show our method outperforms the current best by 1.83% and 1.02% in mIoU and mDSC, respectively.

We present an effective spectral matching method based on a shape association graph for finding region correspondences between two cel animation keyframes. We formulate the correspondence problem as an adapted quadratic assignment problem, which comprehensively considers both the intrinsic geometric and topology of regions to find the globally optimal correspondence. To simultaneously represent the geometric and topological similarities between regions, we propose a shape association graph (SAG), whose node attributes indicate the geometric distance between regions, and whose edge attributes indicate the topological distance between combined region pairs. We convert topological distance to geometric distance between geometric objects with topological features of the pairs, and introduce Kendall shape space to calculate the intrinsic geometric distance. By utilizing the spectral properties of the affinity matrix induced by the SAG, our approach can efficiently extract globally optimal region correspondences, even if shapes have inconsistent topology and severe deformation. It is also robust to shapes undergoing similarity transformations, and compatible with parallel computing techniques.

We present an automated treadmill featuring physiological-data-driven feedback-PhysioTreadmill, which allows its user to easily control running settings based on their physical condition and can also motivate them through real-time physiological computing. We developed a robust exercise intensity self-adaptive adjustment algorithm using physiological data processing to adjust the user's physiological state accurately. We also designed exergames with physiological-data-driven visual/audio feedback in PhysioTreadmill. With PhysioTreadmill, we can ideally increase exercise duration, and enhance exercise performance and safety. Two user studies involving 42 participants showed that PhysioTreadmill is user-friendly and can effectively extend users' training duration.

Flexible indoor scene synthesis is a popular topic in computer graphics and virtual reality research due to its wide-ranging applications in home design, games and automated robotics training. We propose a novel approach to automatic and flexible indoor scene synthesis using an energy-based method. We regard indoor scene synthesis as a multiple-object optimization problem with furniture location and orientation according to the user's intention, as a constraint on the energy of the optimization problem. Based on the relationship of objects, the embedded aesthetic criterion, the design criterion for proper placement and human movement in a scene, we design five energy functions, the overlap constraint, pairwise constraint, wall constraint, aisle constraint, angle constraint and penalty item, are proposed. We use a multi-object particle swarm intelligence optimization method with a Markov chain Monte Carlo algorithm to solve this optimization problem and obtain a Pareto-optimal solution. 3D gestures are used as the medium of interaction between the user and the system. Our method significantly enhances the existing weighted energy optimization method by allowing a joint optimization of various energy functions. The experiments confirm that all the energy functions can converge at the same time and that the proposed method obtains results superior to those of the weighted methods. The proposed method is general which can be used to obtain layouts for various kind of rooms with different furniture.

Abstract Team formation with realistic crowd simulation behavior is a challenge in computer graphics, multiagent control, and social simulation. In this study, we propose a framework of crowd formation via hierarchical planning, which includes cooperative‐task, coordinated‐behavior, and action‐control planning. In cooperative‐task planning, we improve the grid potential field to achieve global path planning for a team. In coordinated‐behavior planning, we propose a time–space table to arrange behavior scheduling for a movement. In action‐control planning, we combine the gaze‐movement angle model and fuzzy logic control to achieve agent action. Our method has several advantages. (1) The hierarchical architecture is guaranteed to match the human decision process from high to low intelligence. (2) The agent plans his behavior only with the local information of his neighbor; the global intelligence of the group emerges from these local interactions. (3) The time–space table fully utilizes three‐dimensional information. Our method is verified using crowds of various densities, from sparse to dense, employing quantitative performance measures. The approach is independent of the simulation model and can be extended to other crowd simulation tasks.