This work develops a novel solution for generating binary patterns of large-scale fabrics using deep neural networks. It contributes to textile engineering by enabling the analysis of ancient and modern textile products. There are only two possible over-under relationships between warp and weft yarns at each crossing point, which can be simulated by a binary matrix. Generating a binary pattern from an observed fabric pattern can help designers save time and effort in reproducing fabrics. Deep neural networks have recently been applied in this field and can generate accurate binary patterns that match fabrics; however, these methods still require improvements. This paper introduces a feature point matching-based image stitching method to address the mismatch between the resolution of large fabric images and the network input requirements. Then, we preserve the contrast of color fabric patterns using principal component analysis for grayscale conversion. Finally, we propose a method for deriving pixel-wise confidence values of the label image based on receptive field size and stitching label images by accumulating weights. We show results on Jacquard fabric samples with an average of 266 thousand intersections. The ablation study showed that incorporating the two newly proposed methods achieved the highest accuracy for the textile binary pattern, with an average of 0.952 across samples.

We introduce a new offset meshing method that handles clean 3D surface meshes of arbitrary geometry and topology—where “clean” refers to meshes that are watertight, manifold, and free of self-intersections. Our approach also extends to imperfect, or “dirty,” meshes that violate these conditions, although the problem becomes significantly more difficult in such scenarios, and faithful feature preservation near defective areas cannot always be assured. In contrast to prior techniques, which have largely focused on constant-radius offsets, our method is, to our knowledge, the first to support mitered offsets while effectively preserving sharp features. Our method is designed based on several core principles: (1) explicitly generating the offset vertices and triangles with feature-capturing energy and constraints; (2) prioritizing the generation of the offset geometry before establishing its connectivity, (3) employing exact algorithms in critical pipeline steps for robustness, balancing the use of floating-point computations for efficiency, (4) applying various conservative speed up strategies including early reject non-contributing computations to the final output. Our approach further uniquely supports variable offset distances on input surface elements, offering a wider range of practical applications compared to conventional methods. For benchmarking purposes, we performed an extensive comparison against state-of-the-art offset methods using a curated subset of the Thingi10K dataset. Our results demonstrate the superiority of our approach over current state-of-the-art methods in terms of element count, feature preservation, and non-uniform offset distances of the resulting offset mesh surfaces, marking a significant advancement in the field.

Occlusion culling is a cornerstone of real-time rendering, particularly in mobile games where limited GPU bandwidth demands highly efficient scene management. At the heart of occlusion culling lies the use of simplified proxy geometry-called occluders-that approximate scene geometry for rapid visibility testing. However, producing high-quality occluders that are low in polygon count, conservative in coverage, and tightly aligned with the original geometry remains a manual and labor-intensive process. In this paper, we present a fast and fully automated two-stage approach for robust occluder generation tailored to real-world game assets. Our method begins with a novel strategy for inward offset mesh computation, followed by a conservative simplification step leveraging a new variant of Quadric Error Metrics (QEM). This approach effectively handles noisy and topologically complex inputs, generating production-ready occluders in seconds. Extensive experiments on a wide range of asset types demonstrate that our technique achieves aggressive triangle reduction while preserving critical occlusion fidelity. By offering a practical and scalable solution, our method bridges the gap between academic research and demanding needs for game development.

Shadows can originate from occlusions in both direct and indirect illumination. Although most current shadow removal research focuses on shadows caused by direct illumination, shadows from indirect illumination are often just as pervasive, particularly in indoor scenes. A significant challenge in removing shadows from indirect illumination is obtaining shadow-free images to train the shadow removal network. To overcome this challenge, we propose a novel rendering pipeline for generating shadowed and shadow-free images under direct and indirect illumination, and create a comprehensive synthetic dataset that contains over 30,000 image pairs, covering various object types and lighting conditions. We also propose an innovative shadow removal network that explicitly integrates semantic and geometric priors through concatenation and attention mechanisms. The experiments show that our method outperforms state-of-the-art shadow removal techniques and can effectively generalize to indoor and outdoor scenes under various lighting conditions, enhancing the overall effectiveness and applicability of shadow removal methods.

Achieving high-quality shadow removal with strong generalizability is challenging in scenes with complex global illumination. Due to the limited diversity in shadow removal datasets, current methods are prone to overfitting training data, often leading to reduced performance on unseen cases. To address this, we leverage the rich visual priors of a pre-trained Stable Diffusion (SD) model and propose a two-stage fine-tuning pipeline to adapt the SD model for stable and efficient shadow removal. In the first stage, we fix the VAE and fine-tune the denoiser in latent space, which yields substantial shadow removal but may lose some high-frequency details. To resolve this, we introduce a second stage, called the detail injection stage. This stage selectively extracts features from the VAE encoder to modulate the decoder, injecting fine details into the final results. Experimental results show that our method outperforms state-of-the-art shadow removal techniques. The cross-dataset evaluation further demonstrates that our method generalizes effectively to unseen data, enhancing the applicability of shadow removal methods.

Volumetric parameterization provides the critical bridge between computer-aided design (CAD) models and Isogeometric Analysis (IGA), where the quality of the parameterization is a primary determinant of computational efficiency and accuracy. Sweeping is a fundamental technique for generating 3D models, and thus, performing volumetric parameterization for sweep-based solids is essential. However, existing sweeping-based parameterization methods are not suitable for multi-axis swept volumes, which are commonly encountered in complex engineering models. To address this limitation, this research introduces an “Decomposition-Parameterization-Recombination” framework. The framework first recursively partitions a complex model into a set of single-axis swept sub-volumes. Subsequently, it generates a high-quality parameterization for each sub-volume by projecting its 3D geometry onto a 2D plane and leveraging frame-field-driven techniques. Finally, through control-point matching and transition zone optimization, these discrete sub-volumes are seamlessly integrated into a globally C0-continuous B-spline volume. Representative case studies validate the proposed framework's capability to effectively generate geometrically accurate volumetric parameterizations for complex multi-axis swept volumes, yielding models that are readily applicable to high-fidelity Isogeometric Analysis.

Integrating pathological image and multimodal information such as patient metadata can help improve the segmentation performance. Most existing segmentation methods overlook the importance to exploit patient metadata, and suffer from suboptimal segmentation performance. To tackle these challenges, a novel multimodal feature fusion module (MFFM) is proposed. By incorporating a cross-modal attention mechanism and a self-attention mechanism, MFFM effectively and efficiently integrates information from different modalities. Experiments are conducted on GlaS, a glandular segmentation dataset, and the experimental results demonstrate that the method outperforms the state-of-the-art segmentation networks.

Medical image segmentation often suffers from performance degradation when applied to images from different domains. To address this, we propose DiMA-Seg (Diffusion Model Adaptation for Segmentation), a novel framework for unsupervised domain adaptation in medical image segmentation. DiMA-Seg combines GAN-based image translation with diffusion model-based feature extraction, leveraging the strengths of both approaches. Our method utilizes the hierarchical nature of diffusion models to extract multi-scale features for accurate segmentation in the target domain. Experiment on MMWHS dataset demonstrates that DiMA-Seg outperforms existing methods in segmentation accuracy.

Micro-actions are spontaneous body movements that indicate a person's true feelings and potential intentions, and micro-action recognition is important in human behavior analysis. Yet, recognizing micro-actions is challenging because they are subtle and appear for a very short time compared to normal actions. In this paper, we propose a micro-action recognition framework based on Hierarchical Fusion and Inference (HiFI) to capture subtle multimodal information. Specifically, we first hierarchically integrate multimodal local and global information, including the 2D key-points of faces, hands and bodies, the depth information, and the RGB image sequences. Afterward, both 3D-CNNs and Transformers are used to effectively capture local and long-range dependence. Finally, we propose a novel from-fine-to-coarse (F2C) inference strategy, based on hybrid ensemble of multi-branches, to boost the accuracy and credibility of coarse action recognition. Our solution ranked 4th in the MAC Challenge Track 1.

3D Human pose estimation from multiple cameras with unknown calibration has received less attention than it should. The few existing data-driven solutions do not fully exploit 3D training data that are available on the market, and typically train from scratch for every novel multi-view scene, which impedes both accuracy and efficiency. We show how to exploit 3D training data to the fullest and associate multiple dynamic views efficiently to achieve high precision on novel scenes using a simple yet effective framework, dubbed Multiple Dynamic View Pose estimation (MDVPose). MDVPose utilizes novel scenarios data to finetune a single-view pretrained motion encoder in multi-view setting, aligns arbitrary number of views in a unified coordinate via Procruste alignment, and imposes multi-view consistency. The proposed method achieves 22.1 mm P-MPJPE or 34.2 mm MPJPE on the challenging in-the-wild Ski-Pose PTZ dataset, which outperforms the state-of-the-art method by 24.8% P-MPJPE (-7.3 mm) and 19.0% MPJPE (-8.0 mm). It also outperforms the state-of-the-art methods by a large margin (-18.2mm P-MPJPE and -28.3mm MPJPE) on the EgoBody dataset. In addition, MDVPose achieves robust performance on the Human3.6M datasets featuring multiple static cameras. Code is available at https://github.com/iGame-Lab/MDVPose.

We present a novel framework for exemplar based image translation. Recent advanced methods for this task mainly focus on establishing cross-domain semantic correspondence, which sequentially dominates image generation in the manner of local style control. Unfortunately, cross-domain semantic matching is challenging; and matching errors ultimately degrade the quality of generated images. To overcome this challenge, we improve the accuracy of matching on the one hand, and diminish the role of matching in image generation on the other hand. To achieve the former, we propose a masked and adaptive transformer (MAT) for learning accurate cross-domain correspondence, and executing context-aware feature augmentation. To achieve the latter, we use source features of the input and global style codes of the exemplar, as sup-plementary information, for decoding an image. Besides, we devise a novel contrastive style learning method, for acquire quality-discriminative style representations, which in turn benefit high-quality image generation. Experimen-tal results show that our method, dubbed MATEBIT, performs considerably better than state-of-the-art methods, in diverse image translation tasks. The codes are available at https://github.com/AiArt-HDU/MATEBIT.

Neural radiance fields (NeRF) based methods have shown amazing performance in synthesizing 3D-consistent photographic images, but fail to generate multi-view portrait drawings. The key is that the basic assumption of these methods -- a surface point is consistent when rendered from different views -- doesn't hold for drawings. In a portrait drawing, the appearance of a facial point may changes when viewed from different angles. Besides, portrait drawings usually present little 3D information and suffer from insufficient training data. To combat this challenge, in this paper, we propose a Semantic-Aware GEnerator (SAGE) for synthesizing multi-view portrait drawings. Our motivation is that facial semantic labels are view-consistent and correlate with drawing techniques. We therefore propose to collaboratively synthesize multi-view semantic maps and the corresponding portrait drawings. To facilitate training, we design a semantic-aware domain translator, which generates portrait drawings based on features of photographic faces. In addition, use data augmentation via synthesis to mitigate collapsed results. We apply SAGE to synthesize multi-view portrait drawings in diverse artistic styles. Experimental results show that SAGE achieves significantly superior or highly competitive performance, compared to existing 3D-aware image synthesis methods. The codes are available at https://github.com/AiArt-HDU/SAGE.

In this paper, a deep learning framework combined with isogeometric analysis (IGA for short) called IGA-Reuse-Net is proposed for efficient reuse of numerical simulation on a set of topology-consistent models. Compared with previous data-driven numerical simulation methods only for simple computational domains, our method can predict high-accuracy PDE solutions over topology-consistent geometries with complex boundaries. UNet3+ architecture with interlaced sparse self-attention (ISSA) module is used to enhance the performance of the network. In addition, we propose a new loss function that combines a coefficients loss and a numerical solution loss. Several training datasets with topology-consistent models are constructed for the proposed framework. To verify the effectiveness of our approach, two different types of Poisson equations with different source functions are solved on three datasets with different topologies. Our framework can achieve a good trade-off between accuracy and efficiency. It outperforms the physics-informed neural network (PINN for short) model and yields promising results of prediction.

A key challenge in the task of human pose and shape estimation is occlusion, including self-occlusions, object-human occlusions, and inter-person occlusions. The lack of diverse and accurate pose and shape training data becomes a major bottleneck, especially for scenes with occlusions in the wild. In this paper, we focus on the estimation of human pose and shape in the case of inter-person occlusions, while also handling object-human occlusions and self-occlusion. We propose a novel framework that synthesizes occlusion-aware silhouette and 2D keypoints data and directly regress to the SMPL pose and shape parameters. A neural 3D mesh renderer is exploited to enable silhouette supervision on the fly, which contributes to great improvements in shape estimation. In addition, keypoints-and-silhouette-driven training data in panoramic viewpoints are synthesized to compensate for the lack of viewpoint diversity in any existing dataset. Experimental results show that we are among the state-of-the-art on the 3DPW and 3DPW-Crowd datasets in terms of pose estimation accuracy. The proposed method evidently outperforms Mesh Transformer, 3DCrowdNet and ROMP in terms of shape estimation. Top performance is also achieved on SSP-3D in terms of shape prediction accuracy. Demo and code will be available at https://igame-lab.github.io/LASOR/.

We propose an angle-based mesh representation, which is invariant under translation, rotation, and uniform scaling, to encode the geometric details of a triangular mesh. Angle-based mesh representation consists of angle quantities defined on the mesh, from which the mesh can be reconstructed uniquely up to translation, rotation, and uniform scaling. The reconstruction process requires solving three sparse linear systems: the first system encodes the length of edges between vertices on the mesh, the second system encodes the relationship of local frames between two adjacent vertices on the mesh, and the third system defines the position of the vertices via the edge length and the local frames. From this angle-based mesh representation, we propose a quasi-angle-preserving mesh deformation system with the least-squares approach via handle translation, rotation, and uniform scaling. Several detail-preserving mesh editing examples are presented to demonstrate the effectiveness of the proposed method.

In isogeometric anlaysis, parameterization of computational domain has great effects as mesh generation in finite element analysis. In this paper, based on the concept of harmonic map from the computational domain to parametric domain, a variational approach is proposed to construct the parameterization of computational domain for 2D isogeometric analysis. Different from the previous elliptic mesh generation method in finite element analysis, the proposed method focus on isogeometric version, and converts the elliptic PDE into a nonlinear optimization problem. A regular term is integrated into the optimization formulation to achieve more uniform grid near convex(concave) parts of the boundary. Several examples are presented to show the efficiency of the proposed method.

The stretching energy functional and the bending energy functional are widely used for approximating the solution of the Plateau-Béizer Problem. This paper presents another two simple methods by using the extended stretching energy functional and the extended bending energy functional. The resulting surface obtained by the new methods will have a smaller area. Comparisons are made with both the area and the mean curvature of the resulting surfaces.

Sculpting deformation is a powerful tool to modify the shape of objects intuitively. However, the detail preserving problem has not been considered in sculpting deformation. In the deformation of a source object by pressing a primitive object against it, the source object is deformed while geometric details of the object should be maintained. In order to address this problem, we present a detail preserving sculpting deformation algorithm by using Laplacian coordinates. Based on the property of Laplacian coordinate, we propose two feature invariants to encode the Laplacian coordinate. Instead of mapping the source mesh to the primitive mesh, we map the smooth version of source mesh to the primitive mesh and use the Laplacian coordinates to encode the geometric details. When the smooth version of the source mesh is deformed, the Laplacian coordinates of the deformed mesh are computed for each vertex firstly and then the deformed mesh is reconstructed by solving a linear system that satisfies the reconstruction of the local details in least squares sense. Several examples are presented to show the effectiveness of the proposed approach.

In this paper, we investigate the direct manipulation problem of free form deformation with rational DMS spline volume (RDMS-FFD). For the weights based direct manipulation method, the solution of the weights can be achieved by solving a linear system; for the control points based method, the explicit solution of displacements of the control points can be obtained, and some properties are also derived from the results. For the constraint points inside the control lattices, we use the weights based methods; for the constraint points outside the control lattices, the control points based method is adopted. Several examples are presented to show the effectiveness of the proposed methods.

Traditional axial deformation is simple and intuitive for users to modify the shape of objects. However, unexpected twist of the object may be obtained. The use of a curve-pair allows the local coordinate frame to be controlled intuitively. However, some important geometric details may be lost and changed in the deformation process. In this paper, we present a detail-preserving axial deformation algorithm based on Laplacian coordinates. Instead of embedding the absolute coordinates into deformation space in traditional axial deformation, we transform the Laplacian coordinates at each vertex according to the transformation of local frames at the closest points on the axial curve. Then the deformed mesh is reconstructed by solving a linear system that describes the reconstruction of the local details in least squares sense. By associating a complex 3D object to a curve-pair, the object can be stretched, bend, twisted intuitively through manipulating the curve-pair, and can also be edited by means of view-dependent sketching. This method combines the advantages of axial deformation and Laplacian mesh editing. Experimental results are presented to show the effectiveness of the proposed method.