【SIGGRAPH 2026 技術論文預告片發放】

全球電腦圖學與互動技術盛會 SIGGRAPH 2026 官方早前釋出本屆 Technical Papers 預告片, 揭示即將於 7 月 19 至 23 日在美國洛杉磯亮相的前沿科技.
今年大會的技術論文核心聚焦於 AI 驅動的工作流程、次世代 Render 技術及動畫算法革新 . 預告片中展示了多項突破性研究, 涵蓋從生成式影片提取神經材質、像素對齊的 3D assets 自動生成到物理驅動的肌肉控制模擬 .
這些先進技術不僅加速了3DCG內容創造效率, 更將互動模擬帶入全新維度, 全面預告了未來 CG 與 AI 融合的行業新標準..

預告中亮點多個技術論文 項目, 來自多間大學, Nvidia , Disney等公司的研發 – 包括 :

– MPM Lite: Linear Kernels and Integration without Particles
– Distributed Affine Body Dynamics with Adaptive Consensus
– JGS2-GQ: Training-free 2nd Jacobi with Gaussian Quadrature
– Mixwell: Sharp 2D Fluid Brushes for Progressive Physics-Based Mixing
– VideoNeuMat: Neural Material Extraction from Generative Video Models
– Toward Richer Material Generation via Procedural Data Enhancement
– BodyReLux: Temporally Consistent Full-Body Video Relighting
– 8DNA: 8D Neural Asset Light Transport by Distribution Learning
– MorphSkein: A Shape-Changing Afterimage Display Preserving Pixel Density During Surface-Area Changes Across Troposkein-Based Shapes
– Volume-Preserving LBM-MPM Coupling for Air-Water-Sand Mixtures
– Interactive Yarn-level Knitwear with Nested Douglas-Rachford Splitting
– Mixed Material Point Methods for Stiff Elastoplasticity
– Fast VEM Fluid Simulation
– Physics-Inspired Procedural Texturing of Extremely Deformable Surfaces
– Pixal3D: Pixel-Aligned 3D Generation from Images
– NeuBase: Spline Surfaces with Neural Basis Functions
– A Few-Step Generative Model on Cumulative Flow Maps
– Computational Design of Coordinate-Motion Assemblies
– AtomSlicer – Constant-Thickness Field-Aligned Non-Planar Slicing and Continuous Toolpaths for FFF
– MOCHI: Motion Enhancement of Collaborative Human-object Interactions
– MotionBricks: Scalable Real-Time Motions with Modular Latent Generative Model and Smart Primitives
– ReActor: Reinforcement Learning for Physics-Aware Motion Retargeting
– Kinematic Kitbashing
– MUSIC: Learning Muscle-Driven Dexterous Hand Control