Dr. Marco Attene -- CNR-IMATI (Genova, Italy)
Marco Attene is a senior researcher at the National Research Council of Italy, where he investigates new directions, paradigms and algorithms for 3D geometric modelling, processing and analysis. He regularly publishes high-impact articles on the most prestigious journals in the area (ACM TOG, TVCG, CGF, …) with thousands of citations, and in 2019 was ranked in the top 2% most influential researchers in all fields of science. His work is regularly presented at all the major computer graphics events including ACM Siggraph and Eurographics, and he is the recipient of the SGP Software Award for his MeshFix software. Marco is an ACM professional member, is vice-president of the Italian Chapter of the Eurographics Association, served as General Chair of the Graphics Replicability Stamp Initiative, and is an associate editor of Wiley Computer Graphics Forum and Elsevier Computers & Graphics.
Robust Algorithms for Reliable Solid Modeling
For researchers and practitioners who deal with geometric models, turning a theoretical algorithm into a reliable computer program might become a tricky and sometimes frustrating task. Even when an algorithm is provably correct, its actual implementation might still fail because it approximates real numbers with finite floating point representations. Exact geometric predicates and adaptive precision may help, but the cost in terms of performance loss can be relevant. This talk analyses and discusses how recent results have tackled this fundamental problem. Starting from Shewchuk's seminal work and CGAL approach to robustness, I will show how geometric algorithms can be made robust with virtually no performance penalty. We will see how the concept of "indirect geometric predicate" has enabled the development of a whole new family of modern algorithms that resolve long lasting problems within the solid and physical modeling community, including 3D arrangements, boolean operations, cascaded editing, volume meshing and multi-material fabrication.
Ye Wang -- Audodesk Research
Ye Wang is a dedicated researcher and entrepreneur, whose journey in technology for design and make has spanned over a decade. She began her research journey at MIT's graphics lab, where she pioneered the development of a volumetric design tool for 3D printing. Building on this foundation, she has been involved in creating successful 3D printing design tools such as Meshmixer and played a key role in the early development of Onshape, the first cloud-based and version-controlled CAD systems. Her entrepreneurial spirit led her to establish a company specializing in collaboration tools for architecture, engineering, and construction. Currently, she serves as a Senior Principal Research Scientist at Autodesk Research, focusing on AI for design knowledge transfer, sustainable design, and conceptual design. She’s also actively engaged in collaborations with academic institutions and participates in industrial research projects aimed at deploying future design tools.
Shaping the Future of Design with Designers: Why is it important to collaborate with industry?
During this keynote, Ye will draw from her experience in design technology development and research to highlight the importance of defining research problems that directly impact design workflows and outcomes. Drawing upon detailed projects such as her recent partnership with automotive companies aimed at bridging design inspiration to concept creation, Ye will champion the necessity of tight collaboration with designers, showcasing how these alliances have profoundly molded her research viewpoints. With the potential of GenAI, collaboration with designers assumes even greater importance. Researchers exploring new design technology must navigate the allocation of responsibilities between designers and AIs with precision. How can we build technology that amplifies human creativity in our research community?
Dr. Jianmin Zheng -- Nanyang Technological University
Dr Jianmin Zheng is a professor in the School of Computer Science and Engineering of Nanyang Technological University. He received his Bachelor degree and Ph.D. degree from Zhejiang University. His research interests include computer aided geometric design, computer graphics, geometric modeling, CAD, Visualization, and interactive digital media. He has done significant research work in his research areas (such as T-spline technology, subdivision surfaces, rational geometric continuity, surface/surface intersection, curve/surface implicitization, and digital media processing algorithms) and published over 200 papers in international journals and conferences such as SIGGRAPH, ACM TOG, IEEE T-PAMI, IEEE TVCG, and IEEE TIP. He has served as a program committee member for many international conferences. Dr Zheng is an associate editor of six journals including "Computer-Aided Design", "The Visual Computer" and "Computers & Graphics".
Reality-Oriented 3D Reconstruction and Shape Parsing
Creating digital 3D models of real-world objects and scenes bridges the gap between the physical and digital worlds, offering significant applications but also presenting substantial challenges. Recent years have seen notable advancements in this area within the fields of computer graphics and vision. This talk begins with an overview of modeling for real-world applications, covering key technical components and the typical demands of such applications. It then presents some of our recent research efforts towards high-level shape representation, and efficient 3D construction and shape parsing that help understand the shapes of real-world objects, support the re-creation of new products, and facilitate practical applications. In particular, we introduce a representation called CSG-Stump, a learning-friendly CSG structure that describes the combination of constituent modeling primitives of a shape in a simple and regular manner. Additionally, we present ExtrudeNet, a network leveraging machine learning to reverse engineer the sketch-and-extrude modeling process of a shape in an unsupervised fashion. We also introduce a differentiable scheme of convex polyhedra to generate convex decompositions of a shape. These methods can be applied to 3D reconstruction and shape parsing from point clouds, multi-view images, and implicit fields.