ACM Transactions on Graphics最新文献

MCHex: marching cubes based adaptive hexahedral mesh generation with guaranteed positive Jacobian MCHex：基于移动立方体的自适应六面体网格生成，保证正雅可比矩阵

IF 6.2 1区计算机科学

ACM Transactions on Graphics Pub Date : 2026-10-13 DOI: 10.1145/3858338

Hua Tong, Yongjie Jessica Zhang

{"title":"MCHex: marching cubes based adaptive hexahedral mesh generation with guaranteed positive Jacobian","authors":"Hua Tong, Yongjie Jessica Zhang","doi":"10.1145/3858338","DOIUrl":"https://doi.org/10.1145/3858338","url":null,"abstract":"Grid-based methods are robust for automatic hexahedral (hex) meshing, but struggle with boundary fidelity and element quality. Conventional pipelines remove outside elements, yielding axis-aligned surfaces converging to the input geometry at first order; subsequent padding and projection are heuristic, offering no guarantees on final boundary fidelity or mesh quality. This paper introduces MCHex, a fundamental reformulation of boundary and mesh-quality handling in grid-based hex meshing. MCHex applies marching cubes (MC) inside each grid cell, ensuring the mesh boundary is an MC surface. The key insight is that constraining cut-cell configurations to 3-regular polyhedra and applying midpoint subdivision produces an all-hex mesh with guaranteed positive Jacobian for every element. MCHex provides three advantages: (1) theoretical positive-Jacobian guarantee for every element; (2) boundary convergence matching MC’s approximation rate, via a non-heuristic algorithm with well-bounded time complexity and the fastest wall-clock time among existing methods; and (3) manifold surfaces for arbitrary geometries plus a natural padded layer. Extensive evaluation on 202 benchmark geometries, using a slightly smaller element budget, compares MCHex with previous state-of-the-art methods; it consistently yields positive-Jacobian meshes with similar boundary fidelity while running significantly faster. It integrates seamlessly with post-processing such as smoothing and simplification, and is suitable for simulation.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"80 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149172508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Design for Machinability: Topology Optimization for Subtractive Machining Using Trimmed Neural Fields 可加工性设计：基于修剪神经场的减法加工拓扑优化

IF 6.2 1区计算机科学

ACM Transactions on Graphics Pub Date : 2026-10-10 DOI: 10.1145/3847303

Hongrui Chen, Levent Burak Kara

{"title":"Design for Machinability: Topology Optimization for Subtractive Machining Using Trimmed Neural Fields","authors":"Hongrui Chen, Levent Burak Kara","doi":"10.1145/3847303","DOIUrl":"https://doi.org/10.1145/3847303","url":null,"abstract":"Topology optimization has found significant utility in additive manufacturing, but the resulting geometries are often difficult to realize using subtractive methods. The complex, curved surfaces of optimized designs typically require time-consuming point milling operations. In contrast, geometries composed of developable surfaces are well suited to flank milling, enabling substantial improvements in machining efficiency and surface quality. However, globally minimizing Gaussian curvature leads to overly simplistic designs. We address this challenge with a novel method that represents geometry using three implicit neural fields, optimized to produce machinable components. The resulting designs consist of regions of developable surfaces separated by seamlines where curvature is concentrated. Seamlines are efficiently computed as intersections of the implicit fields, while the geometry emerges through their product. By constraining surfaces to align with the cutter’s normal or tangential direction, our approach further accommodates practical constraints of 3-axis machining. We demonstrate concurrent optimization of topology and Gaussian curvature, enabling control over seamline feature scale. Compared to methods that enforce subtractive machinability without considering milling efficiency, our approach consistently reduces machining time while maintaining competitive structural performance.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"36 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2026-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149140461","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Class of C 2 Point-Normal Subdivision Surfaces and Their Exact Evaluation at Arbitrary Parameters 一类c2点法线细分曲面及其在任意参数下的精确求值

IF 6.2 1区计算机科学

ACM Transactions on Graphics Pub Date : 2026-09-14 DOI: 10.1145/3847304

Xunnian Yang

{"title":"A Class of C 2 Point-Normal Subdivision Surfaces and Their Exact Evaluation at Arbitrary Parameters","authors":"Xunnian Yang","doi":"10.1145/3847304","DOIUrl":"https://doi.org/10.1145/3847304","url":null,"abstract":"This paper proposes techniques to construct and evaluate C 2 subdivision surfaces under control of quadrilateral meshes with unit normal vectors at the vertices. An improved C 2 modification of the Catmull-Clark subdivision scheme is first given by tuning the eigenvalues of subdivision matrices. Just like the existing C 2 modified Catmull-Clark subdivision, the improved scheme can also generate C 2 subdivision surfaces with vanishing curvature at the extraordinary points, but the flat regions near the extraordinary points have been significantly reduced. By independently applying the new modified subdivision scheme to control points and control normals of the control meshes, a class of C 2 point-normal (PN) subdivision surfaces is defined by updating the limit positions of subdivided points along the limit vectors of subdivided normals. Unlike their linear counterparts, PN subdivision surfaces can approximate control points with even higher-order accuracy while spheres and circular cylinders can be reproduced exactly. The curvatures at the extraordinary points of C 2 PN subdivision surfaces are generally non-vanishing, unless the surfaces reduce to linear subdivision surfaces. Like the classical Catmull-Clark subdivision surfaces, the C 2 modified Catmull-Clark subdivision surfaces and the C 2 PN Catmull-Clark subdivision surfaces can be evaluated exactly at arbitrary parameters. Numerical examples also demonstrate the effectiveness of the proposed techniques for modeling and evaluating curvature-continuous subdivision surfaces.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"50 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2026-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148958005","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Q-DRFormer: Query-Position Mapping Flow Prediction with Progressive Refinement for Document Rectification Q-DRFormer：基于逐级细化的文档纠错查询-位置映射流预测

IF 6.2 1区计算机科学

ACM Transactions on Graphics Pub Date : 2026-09-10 DOI: 10.1145/3845993

Jiajie Li, Jiaxin Zhang, Jianwei Guo, Qiuli Zhang, Xin Zheng, Qian Yin, Wenyi Zeng

{"title":"Q-DRFormer: Query-Position Mapping Flow Prediction with Progressive Refinement for Document Rectification","authors":"Jiajie Li, Jiaxin Zhang, Jianwei Guo, Qiuli Zhang, Xin Zheng, Qian Yin, Wenyi Zeng","doi":"10.1145/3845993","DOIUrl":"https://doi.org/10.1145/3845993","url":null,"abstract":"Document image rectification aims to restore images degraded by geometric distortions. Existing U-shaped encoder-decoder methods rely on hierarchical upsampling operations to recover dense flow fields, which provide limited control over the displacement at individual grid positions, leading to spatial inconsistency and poor detail restoration. To address this issue, we propose Q‑DRFormer, a novel framework that formulates rectification as a set prediction task. Specifically, we design a set of queries, where each query corresponds to a position in the flow field grid and is responsible for regressing a 2D anchor point that indicates the sampling position in the distorted input. This one-to-one design establishes a deterministic query-position mapping, enabling fine-grained control and improving the spatial consistency of the predicted flow field. Exploiting this deterministic correspondence, we further propose a progressive flow refinement strategy that progressively refines anchor points through layer-wise offset prediction to enhance fine-grained restoration. Moreover, considering the lack of downstream-task-oriented evaluation in existing studies, we apply the rectification method to scene table detection and contribute a real-world test set along with corresponding evaluation metrics to fill this gap. Extensive experiments demonstrate that Q-DRFormer outperforms existing methods. The code is available at https://github.com/vvvvvair/Q-DRFormer.","PeriodicalId":50913,"journal":{"name":"ACM Transactions on Graphics","volume":"20 1","pages":""},"PeriodicalIF":6.2,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148922643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Erratum: OmniHands: Robust Motion Capture of Interactive Hands via A Versatile Transformer 勘误：OmniHands：通过多功能变压器交互式手的鲁棒运动捕捉

IF 6.2 1区计算机科学

ACM Transactions on Graphics Pub Date : 2026-08-25 DOI: 10.1145/3839103

Dixuan Lin, Yuxiang Zhang, Mengcheng Li, Wei Jing, Qi Yan, Qianying Wang, Yebin Liu, Hongwen Zhang

引用次数: 0

Erratum: Guidestar-Free Adaptive Optics with Asymmetric Apertures 勘误：导星无自适应光学与不对称孔径

IF 6.2 1区计算机科学

ACM Transactions on Graphics Pub Date : 2026-08-25 DOI: 10.1145/3839110

Weiyun Jiang, Haiyun Guo, Christopher A. Metzler, Ashok Veeraraghavan

引用次数: 0

Erratum: Loops2Roofs: Diffusion-based 3D Roof Generation using a Loop Representation 勘误：Loops2Roofs：使用循环表示的基于扩散的3D屋顶生成