Article
Computer Science, Software Engineering
Kenshi Takayama
Summary: Finding distortion-minimizing homeomorphisms between surfaces of arbitrary genus is a fundamental task in computer graphics and geometry processing. We propose a simple method utilizing intrinsic triangulations to establish consistent images between two models.
ACM TRANSACTIONS ON GRAPHICS
(2022)
Article
Computer Science, Software Engineering
Noam Aigerman, Kunal Gupta, Vladimir G. Kim, Siddhartha Chaudhuri, Jun Saito, Thibault Groueix
Summary: This paper introduces a framework for predicting piecewise linear mappings of arbitrary meshes through neural networks. The framework is able to generate highly detail-preserving maps and is agnostic to the triangulation of the input meshes, allowing for accurate predictions.
ACM TRANSACTIONS ON GRAPHICS
(2022)
Article
Computer Science, Software Engineering
Lin Gao, Tong Wu, Yu-Jie Yuan, Ming-Xian Lin, Yu-Kun Lai, Hao Zhang
Summary: TM-NET is a novel deep generative model that synthesizes textured meshes in a part-aware manner. The model achieves texture compatibility between parts in the same shape through conditional generation and generates high-frequency texture details in a high-dimensional latent space.
ACM TRANSACTIONS ON GRAPHICS
(2021)
Article
Engineering, Electrical & Electronic
Mriganka Sarmah, Arambam Neelima
Summary: Quantization of 3D meshes is important for reducing the number of bits used to represent the vertices. This technique is particularly relevant for dense 3D models, such as medical organ representations, but surface smoothing is a challenge. A novel approach using Bayesian Averaging has been developed to address this challenge.
SIGNAL IMAGE AND VIDEO PROCESSING
(2023)
Article
Computer Science, Artificial Intelligence
Fan Min, Linrong Wang, Shulin Pan, Guojie Song
Summary: This paper proposes a new algorithm that combines deep learning and physical methods for seismic data interpolation. Experimental results show that the proposed method outperforms other methods in terms of visual effects and quantitative evaluation indicators, and it also has interpretability and generalization ability to different seismic data.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Geography, Physical
Weixiao Gao, Liangliang Nan, Bas Boom, Hugo Ledoux
Summary: Recent advancements in data acquisition technology have enabled rapid collection of 3D texture meshes, aiding in urban environment analysis and planning. Semantic segmentation through deep learning enhances understanding but demands a significant amount of labelled data. This research introduces a new benchmark dataset, semi-automatic annotation framework, and annotation tool for 3D meshes, offering potential time savings and comparative analysis for semantic segmentation methods.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Article
Chemistry, Multidisciplinary
Urko Petralanda, Giulia Biffi, Simon C. Boehme, Dmitry Baranov, Roman Krahne, Liberato Manna, Ivan Infante
Summary: Through simulations and experiments, it has been shown that the photoluminescence of Cs4PbBr6 nanocrystals at room temperature is thermally quenched due to strong electron-phonon coupling, shedding light on the controversy surrounding the green emission; investigations indicate that Br vacancies and specific impurities are the main factors contributing to the green emission.
Article
Computer Science, Software Engineering
Gabor Fabian
Summary: In this paper, the well-known Savitzky-Golay filter is extended to three-dimensional triangular meshes for smoothing noisy measurement signals. The original idea of locally polynomial model fitting is naturally adapted for smoothing functions defined on irregular two-dimensional triangular meshes. The focus is on surfaces represented by three-dimensional triangular meshes, which are locally homeomorphic to the two-dimensional topological disk. The generalized filter is defined on the plane, and local embeddings are used to pair the points of the two-dimensional disk and the three-dimensional surface, allowing for fitting low-degree multidimensional polynomials and obtaining a similar continuous model. The method is applied for mesh smoothing and compared to other mesh smoothing methods.
COMPUTER AIDED GEOMETRIC DESIGN
(2023)
Article
Polymer Science
Chen Zou, Hu Zhang, Chen Tan, Zhengguo Cai
Summary: Polyolefins with intrinsic antimicrobial properties were synthesized through direct copolymerization and postpolymerization functionalization routes, resulting in materials with excellent mechanical and antimicrobial properties. Different functional groups play a crucial role in determining the antimicrobial properties of polyolefin materials.
Article
Mathematics, Applied
Michael Quell, Georgios Diamantopoulos, Andreas Hoessinger, Josef Weinbub
Summary: In this study, the multi-mesh fast marching method is extended by a block-based decomposition step to enhance serial and parallel performance on hierarchical meshes. The approach offers improved load balancing with a high mesh partitioning degree, effectively balancing mesh partitions with varying sizes. Various benchmarks and parameter studies are conducted on representative geometries with different complexities, resulting in increased serial performance and achieved parallel speedups on a 24-core Intel Skylake computing platform.
JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS
(2021)
Article
Automation & Control Systems
Zaixing He, Peilong Li, Xinyue Zhao, Shuyou Zhang, Jianrong Tan
Summary: The article presents a method to overcome projection defocus on nonideal surfaces, using an efficient defocus compensation algorithm and estimation method for accurate kernel estimation on complex projection surfaces.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Zhenwei Zhang, Ke Chen, Ke Tang, Yuping Duan
Summary: In this paper, a fast multi-grid algorithm is proposed for minimizing both mean curvature and Gaussian curvature energy functionals without sacrificing accuracy for efficiency. Numerical experiments demonstrate the superiority of this method in preserving geometric structures and fine details.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Huafeng Wang, Xuemei Zhao, Wanquan Liu, Lihong C. Li, Jianhua Ma, Lei Guo
Summary: A dual-domain texture-aware method based on deep learning techniques was proposed in this study for LDCT reconstruction, with the processing of sinogram domain and image domain to achieve better visual effects.
Article
Nanoscience & Nanotechnology
Jiajie Wang, Junmei Zeng
Summary: This paper proposes a virtual reality-based image quality degradation recovery method for nanosensors, which achieves restoration through image denoising and texture complexity analysis. Experimental results demonstrate the effectiveness of the method.
JOURNAL OF BIOMEDICAL NANOTECHNOLOGY
(2021)
Article
Computer Science, Information Systems
Ruijun Ma, Shuyi Li, Bob Zhang, Zhengming Li
Summary: With the development of deep learning technologies, the proposed HPDNet achieves a considerable improvement in performance for real-world noisy image denoising, providing a balanced trade-off between denoising accuracy and efficiency.
IEEE TRANSACTIONS ON MULTIMEDIA
(2022)
Article
Computer Science, Software Engineering
Xinxin Zhang, Yuefeng Xi, Zhentao Huang, Lintao Zheng, Hui Huang, Yueshan Xiong, Kai Xu
Summary: This paper proposes a novel high-accuracy active hand-eye calibration approach that improves the calibration accuracy through guiding robot movement and camera view selection. The method employs an online estimated discrete viewing quality field to guide data acquisition and selects the next-best-view based on view quality. Experimental results show that the algorithm outperforms other approaches in terms of accuracy and robustness.
Article
Robotics
Tan Zhang, Kefang Zhang, Jiatao Lin, Wing-Yue Geoffrey Louie, Hui Huang
Summary: This paper proposes a unified representation for obstacle avoidance of robotic manipulators in unstructured environments using sim-to-real deep reinforcement learning. The unified representation is achieved through a vision-based actor-critic framework with a bounding box predictor module. The end-to-end model of the unified representation outperforms state-of-the-art techniques in sim-to-real adaptation and scene generalization.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Engineering, Civil
Pengdi Huang, Liqiang Lin, Kai Xu, Hui Huang
Summary: Autonomous 3D acquisition of outdoor environments presents unique challenges, requiring both discrete and continuous optimization for energy-efficient scanning. The approach involves computing a topological map and optimizing traverse paths between visit sites.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Computer Science, Software Engineering
Qian Zheng, Min Lu, Sicong Wu, Ruizhen Hu, Joel Lanir, Hui Huang
Summary: This paper proposes an effective method to automatically generate coloring for categorical data visualization, which resembles a reference image while allowing classes to be easily distinguished. The method extracts dominant and discriminable colors from the reference image's color space and optimizes point distinctness and color spatial relations to assign colors to the given classes.
COMPUTATIONAL VISUAL MEDIA
(2022)
Article
Computer Science, Software Engineering
Yuqing Lan, Yao Duan, Chenyi Liu, Chenyang Zhu, Yueshan Xiong, Hui Huang, Kai Xu
Summary: In this paper, we propose a novel 3D attention-based relation module (ARM3D) that extracts object-aware relation contexts and filters out irrelevant or confusing contexts through attention mechanism, thereby improving the accuracy and robustness of 3D object detection.
COMPUTATIONAL VISUAL MEDIA
(2022)
Article
Computer Science, Software Engineering
Ruizhen Hu, Xiangyu Su, Xiangkai Chen, Oliver Van Kaick, Hui Huang
Summary: We propose an automatic method for assigning photorealistic relightable materials to 3D shapes. Our method combines image translation and material prediction neural networks to guide the assignment of materials based on a photo exemplar. The key ideas include establishing a correspondence between the exemplar and shape projection and using the translated images to guide material assignment for consistency.
ACM TRANSACTIONS ON GRAPHICS
(2022)
Article
Computer Science, Software Engineering
Qijin She, Ruizhen Hu, Juzhan Xu, Min Liu, Kai Xu, Hui Huang
Summary: This paper proposes a method of joint planning of grasp and motion using deep reinforcement learning to tackle the problem of high-DOF reaching-and-grasping. By adopting the Interaction Bisector Surface (IBS) as a representation of gripper-object interaction, and with the help of several technical contributions, such as fast IBS approximation, vector-based reward, and effective training strategy, the authors achieve a strong control model of high-DOF grasping with good sample efficiency, dynamic adaptability, and cross-category generality. Experimental results demonstrate that the proposed method can generate high-quality dexterous grasp for complex shapes with smooth grasping motions.
ACM TRANSACTIONS ON GRAPHICS
(2022)
Article
Computer Science, Software Engineering
Min Lu, Noa Fish, Shuaiqi Wang, Joel Lanir, Daniel Cohen-Or, Hui Huang
Summary: This article proposes incorporating data-driven animations into static charts to enhance data encoding and emphasize specific attributes. The impact and effectiveness of the animated effects on visual understanding are evaluated through experiments and user studies.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2022)
Article
Computer Science, Software Engineering
Hui Wang, Bitao Ma, Junjie Cao, Xiuping Liu, Hui Huang
Summary: We propose a novel method that addresses the problem of multiple solutions in deep functional map matching for shapes with left-to-right reflectional intrinsic symmetries. Our method can detect both direct correspondences and symmetric correspondences among shapes simultaneously. It also detects the reflectional intrinsic symmetry of each shape. This is achieved by using two Siamese networks, learning consistent direct descriptors and their symmetric counterparts, combined with carefully designed regularized functional maps and supervised loss.
Article
Computer Science, Software Engineering
Yilin Liu, Liqiang Lin, Yue Hu, Ke Xie, Chi-Wing Fu, Hao Zhang, Hui Huang
Summary: The study introduces a new learning-based reconstructability predictor for large-scale 3D urban scene acquisition using unmanned drones. By predicting scene reconstructability through model construction, drone path planning can be guided more accurately compared to heuristic approaches.
ACM TRANSACTIONS ON GRAPHICS
(2022)
Article
Computer Science, Software Engineering
Pengfei Xu, Yifan Li, Zhijin Yang, Weiran Shi, Hongbo Fu, Hui Huang
Summary: This paper introduces a novel method for blending hierarchical layouts with semantic labels, utilizing a hierarchical structure correspondence algorithm to achieve globally optimal correspondence. The resulting compound structure helps extract intermediate layout structures, while also defining a similarity measure between layouts in a hierarchically structured view.
ACM TRANSACTIONS ON GRAPHICS
(2022)
Article
Engineering, Electrical & Electronic
Qingquan Li, Hui Huang, Wenshuai Yu, San Jiang
Summary: Unmanned aerial vehicles have become widely used in remote sensing and are critical in the construction of smart cities. However, urban environments pose challenges for secure and accurate data acquisition for 3D modeling. This study presents optimized views photogrammetry as a solution and verifies its precision and potential in large-scale 3D modeling.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Computer Science, Information Systems
Liqiang Lin, Pengdi Huang, Chi-Wing Fu, Kai Xu, Hao Zhang, Hui Huang
Summary: We propose a new attention-based mechanism for learning enhanced point features in point cloud processing tasks. Unlike previous studies, our approach learns to locate the best attention points to optimize the performance of specific tasks. We advocate the use of single attention points for better semantic understanding in point feature learning.
SCIENCE CHINA-INFORMATION SCIENCES
(2023)
Article
Computer Science, Software Engineering
Yilin Liu, Ruiqi Cui, Ke Xie, Minglun Gong, Hui Huang
Summary: Traditional urban reconstruction methods can only output incomplete 3D models, while learning-based shape reconstruction techniques are designed for single objects. This paper proposes a novel learning-based approach for real-time complete 3D mesh reconstruction of large-scale urban scenes. The approach segments objects and determines their positions, and reconstructs them under local coordinates to approximate training datasets.
COMPUTERS & GRAPHICS-UK
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Zimu Yi, Ke Xie, Jiahui Lyu, Minglun Gong, Hui Huang
Summary: The use of image-based rendering (IBR) technique is important for implementing VR telepresence by allowing interactive presentation of real scenes to viewers. However, the quality of IBR results depends on various factors such as pre-captured views and rendering algorithms. In this work, we introduce the concept of renderability, which predicts the quality of IBR results at any given viewpoint and view direction, to guide the selection of viewpoints/trajectories for challenging large-scale 3D scenes.
2023 IEEE CONFERENCE VIRTUAL REALITY AND 3D USER INTERFACES, VR
(2023)
