BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
X-LIC-LOCATION:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:19911015T033000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251218T030657Z
LOCATION:Hall 3F/G Concourse\, Level 3
DTSTART;TZID=Asia/Hong_Kong:20251216T090000
DTEND;TZID=Asia/Hong_Kong:20251216T180000
UID:siggraphasia_SIGGRAPH Asia 2025_sess205@linklings.com
SUMMARY:Posters
DESCRIPTION:The Posters program at SIGGRAPH Asia 2025 offers a dynamic pla
 tform for sharing new ideas, early research findings, practical solutions,
  and creative experiments from across the CG and interactive techniques co
 mmunity.\n\nEach poster presents insights spanning a wide range of topics 
 – from animation and rendering to machine learning, XR, virtual worlds, an
 d beyond. It’s an opportunity to discover the latest developments and conn
 ect directly with the authors behind the work.\n\nThis year, we spotlight 
 projects inspired by generative technologies – from human-AI collaboration
  and creative coding to procedural design and AI-driven workflows. Posters
  will showcase how generative systems are influencing not only the tools w
 e use, but the ways we think, create, and communicate.\n\nWhether you’re a
  seasoned expert or a curious newcomer, the Posters program invites you to
  engage, exchange, and be inspired by the ideas shaping tomorrow.\n\nEgoce
 ntric 3D Pose Estimation with Temporally Guided Pose Refinement\n\nWe prop
 ose a two-stage pipeline combining causal TCN and Transformer refinement, 
 enabling accurate and temporally coherent 3D egocentric pose estimation un
 der severe occlusions with minimal latency.\n\n\nSeongmin Baek, Youn-Hee G
 il, and Hyunjun Lim (Electronics and Telecommunications Research Institute
  (ETRI))\n---------------------\nPolyArt: Customizable Multilingual Movie 
 Poster Generation via Diffusion Transformer\n\nCustomizable multilingual a
 esthetic movie poster generator for diverse audiences.\n\n\nXin Fu (Tsingh
 ua University); Nisha Huang (Tsinghua University, Pengcheng Laboratory); J
 unjie Li (Tsinghua University); Jie Guo (Pengcheng Laboratory); Xiu Li (Ts
 inghua University); and Tong-Yee Lee (National Cheng-Kung University)\n---
 ------------------\nDrawing Lines to Connect: A Multi-Participant XR Audio
 visual Art Experience Using Internet-Connected Video See-Through MR with E
 EG/PPG\n\nIn this internet-connected MR art experience, participants co-cr
 eate by drawing lines in mid-air, generating sounds and VR landscapes, whi
 le estimated emotions are reflected as shifting colors in avatars and line
 s.\n\n\nRyu Nakagawa (Nagoya City University); Hidefumi Ohmura (Tokyo Univ
 ersity of Science); and Kenta Hidaka, Sho Kato, and Ai Kawaguchi (Nagoya C
 ity University)\n---------------------\nGraph Q-Learning for Automatic Ass
 embly in Design-to-Construction\n\nWe propose a CAD2CAM system using Graph
  Reinforcement Learning for brick assembly. Our method optimizes assembly 
 stability, reducing costs and enhancing structural integrity in mobile man
 ipulator tasks.\n\n\nShih-Yu Lai (National Taiwan University), Chia-Ching 
 Yen and Yang-Ting Shen (National Cheng Kung University), and Bing-Yu Chen 
 (National Taiwan University)\n---------------------\nBamboo Rhyme Craft: I
 mmersive VR Bamboo Weaving through Coordinated Bimanual Gesture Interactio
 n\n\nImmersive VR with coordinated bimanual gesture interaction revives tr
 aditional bamboo weaving, blending cultural heritage with engaging, hands-
 on experiences to promote, preserve, and creatively reinterpret intangible
  traditions for wider audiences.\n\n\nChenyue Zheng, Yaxuan Zhao, and Runx
 in Guo (School of New Media Art and Design, Beihang University) and Shuo Y
 an (School of New Media Art and Design,Beihang University; State Key Labor
 atory of Virtual Reality Technology and Systems, Beihang University)\n----
 -----------------\nCalligraphy Echo: Sonifying Chinese Calligraphy with MI
 DI-Enabled Brush\n\nCalligraphy Echo is a MIDI-enabled brush that translat
 es real-time calligraphy dynamics into traditional instrument sounds, maki
 ng the art's essence audible.\n\n\nYizheng Xu (University of Waikato Insti
 tute(NZUWI), Hangzhou City University); Kexin Nie (The University of Sydne
 y); Yufeng Wu (University of Technology Sydney); Huiying Lu (Harbin Instit
 ute of Technology); and Longyu Zhang (Zhejiang University)\n--------------
 -------\nHigh-Clarity Underwater 3D Display Enabled by Electrolysis-Free W
 ater-Mediated Wireless Power Transmission\n\nWe propose an electrolysis-fr
 ee underwater wireless power transfer method enabling clear bubble-free 3D
  displays. The method powers 7 mm cubic devices and supports 115.2 kbps co
 mmunication, achieving high-clarity volumetric visualization.\n\n\nChiaki 
 Takahashi and Hayato Sawada (Rakusei High School) and Ryo Shirai and Masan
 ori Hashimoto (Kyoto University)\n---------------------\nRingFlowUI: 3D In
 terface System using Ring-mounted Camera\n\nRingFlowUI uses a finger-mount
 ed camera and deep learning to infer 3D hand shape and motion, enabling na
 tural, fatigue-free VR/AR interaction with compact, intuitive, and immersi
 ve hand-based control.\n\n\nMatthew Ichikawa, Saori Takeyama, and Masahiro
  Yamaguchi (Institute of Science Tokyo)\n---------------------\nToward Mob
 ile 3D Light Field Displays: An Efficient 3DGS Rendering Framework\n\nThis
  paper presents an optimized 3D Gaussian Splatting framework for real-time
  light field rendering on mobile devices, demonstrating the feasibility of
  portable and immersive 3D experiences anywhere.\n\n\nHong-Chang Shin, Hos
 ung Jeon, Joonsoo Kim, Gwangsoon Lee, and HyungGon Choo (Electronics and T
 elecommunications Research Institute (ETRI))\n---------------------\nAmpli
 tude-Guided Phase Masking for Hologram Compression\n\nWe introduce an ampl
 itude-guided phase masking strategy for hologram compression as a pre-proc
 essing step for existing image coding. It selectively preserves phase info
 rmation in high optical intensity regions to improve coding efficiency. Th
 e method maintains compatibility with widely used encoders...\n\n\nHiroki 
 Kojima (KDDI Research, Inc.) and Haruhisa Kato and Keisuke Nonaka (KDDI Re
 search, inc.)\n---------------------\nkoloHart: Neck-Worn Tactile Choker f
 or Sensory Substitution and Sensory Enhancement\n\nTactile feedback on the
  neck is considered to hold great potential due to its unique sensory char
 acteristics as a sensory organ. However, this area remains largely unexplo
 red at present.\nTo address this potential, we propose koloHart, a neck-wo
 rn tactile choker designed to utilize the neck's distinc...\n\n\nTakafumi 
 Morita, Haruki Momiyama, and Shogo Tochimoto (The University of Tokyo)\n--
 -------------------\nEffectiveness of "Focus Lines" for Attentional Guidan
 ce in Virtual Reality\n\nThis study proposes applying manga-style "focus l
 ines" for attentional guidance in VR , demonstrating through comparative e
 xperiments that they can enhance guidance effects over conventional method
 s while suggesting it is possible to maintain immersion by adjusting param
 eters.\n\n\nKenji Tomita and Mie Sato (Utsunomiya University)\n-----------
 ----------\nKomorebi-Mediated Ambient Notifications for Calm Technology\n\
 nKomorebi-mediated ambient notifications use dappled light to minimize att
 entional capture. Mechanical and optical prototypes steer movement and sha
 dow hardness, offering gentle cues for time, status, and personal prompts 
 in spaces.\n\n\nMamoru Hiramatsu and Toshitaka Amaoka (Meisei University)\
 n---------------------\nImproving Angular Parameterization for Compact Neu
 ral Materials\n\nWe study MLP input angular parameterizations' impact on c
 ompact neural materials, showing suitable choices improve visual quality, 
 especially for small MLPs, supported by extensive experiments and practica
 l recommendations.\n\n\nZilin Xu (Mohamed bin Zayed University of Artifici
 al Intelligence, Meta Reality Labs Research); Yang Zhou and Yehonathan Lit
 man (Meta Reality Labs Research); Lingqi Yan (Mohamed bin Zayed University
  of Artificial Intelligence); and Anton Michels (Meta Reality Labs Researc
 h)\n---------------------\nLineDiff: Semi-generative Line-drawing Frame In
 terpolation\n\nLineDiff explores the possibility of incorporating a genera
 tive optical flow network into a traditional warp-and-synthesis frame inte
 rpolation network for line-drawing data, with the aim of embracing the mer
 its of both.\n\n\nLiang Kong and Suguru Saito (Institute of Science Tokyo)
 \n---------------------\nHaptiX: A Fusion Framework for Spatiotemporal Cro
 ss-Modal Haptic Synthesis on Mobile Devices\n\nWe present Haptix - a frame
 work for context-aware haptics synthesis for mobile cinematic videos, enha
 ncing immersion through spatiotemporal alignment. User studies highlight e
 ffectiveness, adaptive design needs, and immersive experiences.\n\n\nVeera
 manohar Avudaiappan, Sakthivel Saravanathangam, and Nithiesh P. Natarajan 
 (Department of Electrical and Electronics Engineering, Amrita School of En
 gineering, Coimbatore, Amrita Vishwa Vidyapeetham) and Ritwik Murali (Depa
 rtment of Computer Science and Engineering, Amrita School of Computing, Co
 imbatore, Amrita Vishwa Vidyapeetham)\n---------------------\nA Hybrid XR 
 Framework for Collaborative Scientific Visualization: Rendering-as-a-Servi
 ce with ParaView and Real-Time Engines\n\nHybrid RaaS architecture couples
  ParaView/VTK with XR clients for low-latency, multi-user collaboration; e
 xplicit server-driven sync and intent-only control sustain interactive rat
 es and coherence on TB–PB datasets across heterogeneous devices.\n\n\nJin 
 Sung Choi (Electronics and Telecommunications Research Institute (ETRI))\n
 ---------------------\nFlying in VR: A Therapist-Supervised Prototype for 
 Aviophobia\n\nTherapist-supervised VR flight-exposure prototype pairs Meta
  Quest Pro with a three-actuator seat and web dashboard. Feasibility resul
 ts: high usability and engagement, minimal simulator sickness, no adverse 
 events, supporting aviophobia research.\n\n\nSebastian Cieślik, Juliusz Zo
 lich, Michał Colautti, Eryk Klemencki, Igor Frątczak, Grzegorz Zwoliński, 
 and Dorota Kamińska (Lodz University of Technology)\n---------------------
 \nGenerative POPs: Automating Character-Based Promotions on Digital Signag
 e in Retail Stores\n\nWe propose a product recommendation system that util
 izes generative AI to automatically create characters based on product fea
 tures extracted from product images and instantly display them on shelf-ed
 ge digital signage.\n\nIn a user study with six store staff members, the s
 ystem was highly evaluated f...\n\n\nGen Ono (Shibaura Institute of Techno
 logy); Takuya Iwamoto and Kojiro Tanaka (CyberAgent,inc); and Soh Masuko (
 Shibaura Institute of Technology)\n---------------------\nDynamic Eye High
 light Deformation via Soft-body Simulation in 3D Anime Character Rendering
 \n\nOur method stylizes eye highlights in 3D anime-style characters by int
 roducing Position-Based Dynamics into reflection mapping and solving it as
  a constraint.\n\n\nAtsuki Haruyama and Yuki Morimoto (Kyushu University)\
 n---------------------\nTangible Immersive Authoring of Dynamic Mixed Real
 ity Scenes with the World in Miniature Paradigm\n\nWe present work-in-prog
 ress on the development of a novel framework employing the World-in-Miniat
 ure paradigm, Tangible Interaction and Space-time bounds for efficient and
  intuitive immersive authoring of dynamic Mixed Reality scenes.\n\n\nXuyu 
 Li and John Dingliana (Trinity College Dublin)\n---------------------\nHyb
 ridDeform4D: Joint Refinement of Mesh and Gaussian Splatting for Video-to-
 4D Object Generation\n\nBy integrating the joint refinement of Mesh and Ga
 ussian Splatting, HybridDeform4D captures complex motion and appearance va
 riations in video-to-4D object generation, improving temporal dynamics wit
 hout requiring high-quality initialization.\n\n\nDuotun Wang (The Hong Kon
 g University of Science and Technology (Guangzhou)); Yuhang Li (Lenovo CTO
 O); Zhijing Shao (The Hong Kong University of Science and Technology (Guan
 gzhou)); Wanjun Lv and Liuxin Zhang (Lenovo CTOO); and Mingming Fan and Ze
 yu Wang (The Hong Kong University of Science and Technology (Guangzhou), T
 he Hong Kong University of Science and Technology)\n---------------------\
 nShen Shu Yu Jin Wu: XR Virtual Production Empowering the Visual Spatial C
 onstruction and Immersive Experience of Chinese Classical Dance Films\n\nT
 he XR-based film Shen Shu Yu Jin Wu draws inspiration from the ancient Chi
 nese myth of the three-legged golden crow in The Classic of Mountains and 
 Seas: Eastern Sea Chapter.\n\n\nLing Zou (Beihang University; State Key La
 boratory of Virtual Reality Technology and Systems, Beihang University); Y
 ijia Li (University of Arts London); Richeng Liu (Abertay University); and
  Xizi Liu (Beihang University)\n---------------------\nThread Display: A N
 ovel Wind-Driven Spatial Display Technique\n\nOur research presents the Th
 read Display: a novel screen of wind-suspended, retro-reflective threads. 
  Dynamically shaping the viewing area for a truly unique visual. This retr
 action capability minimizes visual intrusion and enhances the floating eff
 ect. By eliminating material replenishment, our me...\n\n\nTsukasa Ebisawa
  and Naoki Hashimoto (The University of Electro-Communications)\n---------
 ------------\nEfficient Visualization of Large-Scale Unstructured Grid Dat
 a through Importance-Based Hybrid Reduction\n\nFor large-scale CFD/FEA gri
 ds, we propose importance-driven hybrid reduction—explicit decimation for 
 low-importance and learning-based reduction for salient regions—achieving 
 3–8× size reduction with ≈99% accuracy and higher FPS.\n\n\nYongwan Kim (E
 lectronics and Telecommunications Research Institute (ETRI))\n------------
 ---------\nThe Theater Machine: An Interactive Generative-AI-Based Install
 ation for Participatory Cultural Mediation\n\nThe Theater Machine is an in
 teractive AI-based installation designed for theater foyers to make perfor
 ming arts more accessible through participatory engagement. Using camera a
 nd touchscreen interaction, visitors are integrated in real time into dyna
 mically generated theatrical scenes. The system op...\n\n\nNils Gallist, J
 olanda Abasolo, and Celine Pham (University of Applied Sciences Upper Aust
 ria) and Juergen Hagler (University of Applied Sciences Upper Austria, Ars
  Electronica Australia)\n---------------------\nAural Fields: A Real-Time 
 Spatial Audio System Using Grid-Based Probes and Occlusion-Aware Attenuati
 on\n\nAural Fields: A Novel Real-Time Spatial Audio System Using Grid-Base
 d Probes and Raytraced Occlusion-Aware Attenuation.\n\n\nRaghav Suriyashek
 ar and Vasileios Argyriou (Kingston University)\n---------------------\nTr
 yIto: Discrete IMEX Cloth Simulation with Learned Residual Forces\n\nThis 
 poster presents TryIto, a hybrid cloth simulation framework combining thin
 -shell mechanics, learned residual forces, and a dynamic 4D UV atlas, enab
 ling more stable, accurate, and realistic virtual garment simulation.\n\n\
 nAva Oppenheimer (TryIto) and Sebastian Marino (Love & Science Inc.)\n----
 -----------------\nEngine-Integrated Cel-Shading for Unreal Engine 5\n\nTh
 is study aims to implement a cel-shader in Unreal Engine and discuss its p
 ros and cons, such as limitations and how one may be able to do so.\n\n\nG
 abriel Lucena and Cesar Rennó-Costa (AKCIT-IMD, Universidade Federal do Ri
 o Grande do Norte (UFRN)); Charles Madeira (Universidade Federal do Rio Gr
 ande do Norte (UFRN)); and Alyson Souza (AKCIT-IMD, Universidade Federal d
 o Rio Grande do Norte (UFRN))\n---------------------\nMindscape: Multi-Per
 spective VR Storytelling in Interactive Psychological Thriller Experiences
 \n\nMindscape presents a VR psychological thriller where players embody a 
 psychiatrist investigating murders while discovering their fragmented iden
 tity through multi-perspective storytelling, environmental symbolism, and 
 innovative interaction design mechanics.\n\n\nZizhuo Wang, Ziyan Xiong, Ji
 ayu Zhang, Ranran Zeng, Yu Fang, Guanyou Li, and Mengyao Guo (Harbin Insti
 tute of Technology, Shenzhen); Pei Zhang (Shenzhen Polytechnic University)
 ; and Zhiwei Deng (Configreality Technology Co.Ltd.)\n--------------------
 -\nFisheye Patch-Based Automatic Colorization Method for Anime Line-drawin
 gs\n\nOur fisheye patch structure efficiently incorporates global context 
 in patch-based learning by compressing surroundings while preserving centr
 al detail, achieving accurate and consistent colorization of anime line dr
 awings.\n\n\nRyosuke Takamura and Taiki Watanabe (Chiba University); Akino
 bu Maejima (OLM Digital, Inc. IMAGICA GROUP Inc.); Seitaro Shinagawa (Nara
  Institute of Science and Technology (NAIST)); Takuya Funatomi (Kyoto Univ
 ersity, Nara Institute of Science and Technology); Yasuhiro Mukaigawa (Nar
 a Institute of Science and Technology (NAIST)); Shigeo Morishima (Waseda U
 niversity); and Hiroyuki Kubo (Chiba University)\n---------------------\nI
 mmersive Brain Exploration: A VR-Based Gamified System for Neuroanatomy Ed
 ucation with AI-Augmented Learning\n\nA VR neuroanatomy system transforms 
 medical education through immersive 3D brain visualization, AI-generated v
 isual mnemonics, and gamified learning experiences that significantly enha
 nce students' spatial understanding and knowledge retention.\n\n\nQingzhu 
 Zhang (University of California Berkeley)\n---------------------\nForce Fe
 edback Button that Focuses on Both Operation and Non-Operation\n\nWe propo
 se a force feedback button, which can provide three types of feedback: "pr
 essing force," "click feeling," and "stroke." Using this feedback, we deve
 loped application examples for various situations.\n\n\nKazuma Tahara and 
 Koji Tsukada (Future University Hakodate)\n---------------------\nReimagin
 ing the CGI Pipeline with Diffusion Models: A Renderless Workflow for 3D P
 roduct Visualization\n\nThis modular, artist-guided diffusion workflow rei
 magines CGI pipelines by generating 3D product videos from images in minut
 es, drastically reducing production times and enabling designers to iterat
 e faster with creative freedom.\n\n\nAthirah Zaini (University of Auckland
 ), Auzaei Saffiey (Fisher & Paykel Appliances Limited), and Allan Fowler a
 nd Xue Bai (University of Auckland)\n---------------------\nSWEAR: A Multi
 modal Dataset and Embodied Emotion Simulation System for Understanding Hyp
 erhidrosis through Interactive Wearable Jewellery\n\nSWEAR is a bio-intera
 ctive jewelry system combining PPG sensing and sweat simulation to evoke e
 mbodied empathy. Through multimodal, stress-responsive interventions, it e
 nables users to experience hyperhidrosis symptoms, bridging affective comp
 uting and wearable design for impactful applications in h...\n\n\nYin Chan
 g (Shanghai Jian Qiao University), Xiyang Chen (School of art institute of
  Chicago), Tairan Sang (Central Saint Martins College of Art and Design), 
 and Ze Gao (Dawa AIGC Research Institute)\n---------------------\nSpatial 
 Audio for Ski Simulation with Visually Impaired Users in Dynamic Environme
 nts\n\nWe propose a skiing system for people with visual impairments (PVI)
  that uses spatial audio to indicate the positions of obstacles in dynamic
  environments.\n\n\nYuta Taguchi (Waseda University); Toshihiro Hirano (In
 stitute of Science Tokyo); Masaki Kuribayashi (Waseda University); Yichen 
 Peng, Erwin Wu, and Hideki Koike (Institute of Science Tokyo); and Shigeo 
 Morishima (Waseda Research Institute for Science and Engineering)\n-------
 --------------\nVisualizing the Spatio-Temporal Narratives of Hong Kong’s 
 Southbound Literati\n\nNanlai Zhi Hua visualizes the spatio-temporal narra
 tives of Hong Kong’s Southbound Literati through “data flowers” that merge
  bio- graphical and fictional geographies. Integrating literary geography 
 with information visualization, the project translates displacement and no
 stalgia i...\n\n\nYuan Zhang and Angela Chulei Tang (Tsinghua University; 
 Department of Information Art & Design, Academy of Arts & Design); Yan son
 g Chen (Beijing Normal University, School of Arts & Communication); Yi Yi 
 Qian (China Academy of Art, Department of Digital Media Art); WenBo Wang a
 nd Min zhi Shen: (Tsinghua University, Department of Information Art & Des
 ign); and Qi Duanmu and Yin Bing (China Academy of Art, School of Design a
 nd Innovation)\n---------------------\nOuroboros Code: A Practice-Based St
 udy on Sci-Fi Atmosphere Design under Nonlinear Narrative Structures\n\nOu
 roboros Code shows how color, lighting, and spatial design transform nonli
 near science fiction into clear, emotionally resonant worlds.\n\nValidated
  by literature, case studies, and playtests, the researchers present a fra
 mework.\n\n\nXianzhe Long and Zhiyu Zhao (Beihang University)\n-----------
 ----------\nEnhancing Embodiment and Spatial Awareness in VR with Hybrid V
 isual Presentation via Binocular Rivalry\n\nWe propose a novel VR method c
 ombining first-person and third-person perspectives. It uses binocular riv
 alry to balance embodied sensation and spatial awareness. Visual enhanceme
 nts improve the effect and task efficiency.\n\n\nHironobu Sakai and Naoki 
 Hashimoto (The University of Electro-Communications)\n--------------------
 -\nDesignEval: A Human-Centered Framework for Aligning Graphic Designs wit
 h User Intent through Quantitative and Multimodal Evaluation\n\nDesignEval
  is a human-centered framework integrating vision, language, and multimoda
 l models to evaluate graphic designs against user goals, providing quantit
 ative, interpretable feedback, and enhancing creativity, workflow efficien
 cy, and stakeholder communication.\n\n\nVeeramanohar Avudaiappan (Departme
 nt of Electrical and Electronics Engineering, Amrita School of Engineering
 , Coimbatore, Amrita Vishwa Vidyapeetham); Dilip Parasu (Department of Art
 ificial Intelligence, Amrita School of Engineering, Coimbatore, Amrita Vis
 hwa Vidyapeetham); and Ritwik Murali (Department of Computer Science and E
 ngineering, Amrita School of Computing, Coimbatore, Amrita Vishwa Vidyapee
 tham)\n---------------------\nAnime Face Alignment via Guideline-Based Tra
 nsform for Expression Analysis\n\nWe present a guideline-based canonicaliz
 ation of anime-style faces using sparse feature lines, enabling pose-invar
 iant alignment, expression clustering, and transfer for stylized facial an
 alysis and design support.\n\n\nHideki Todo, Ryosuke Higasayama, Sho Harag
 uchi, and Shintaro Horikoshi (Takushoku University)\n---------------------
 \nInteractive generation and stereoscopic observation of 3DCG scenes based
  on 2D tabletop drawing\n\nWe propose a novel drawing system that integrat
 es the entire process of interactive generating and observing 3DCG scenes 
 on the tabletop from 2D sketches.\n\n\nAzumi Katayama and Shinji Mizuno (A
 ichi Institute of Technology) and Kenji Funahashi (Nagoya Institute of Tec
 hnology)\n---------------------\nEchoes of Poetry: Recontextualizing Class
 ical Chinese Poetry into Modern Pop Music through Narrative Visualization\
 n\nThe project recontextualizes classical Chinese poetry into modern pop m
 usic through interactive narrative visualization, blending tradition with 
 modernity to reveal new meanings, emotions, and cultural functions of Chin
 ese poetry.\n\n\nYechun Peng, Xin Chen, Dianheng Jiang, Hanyu Dong, Nan Ca
 o, and Yang Shi (Tongji University, Intelligent Big Data Visualization Lab
 )\n---------------------\nComparing 3D Data Representations for Keypoint E
 stimation in Humanoid Characters\n\nThis study compares voxel, point cloud
 , and image-based 3D data representations for humanoid keypoint estimation
  using geometric deep learning, with a rigger study revealing diverse prac
 tices and subjective rigging standards.\n\n\nJeanne-Emma Lefèvre (Kinetix,
  Université Paris Dauphine-PSL); Théo Cheynel (Kinetix, Centre National de
  la Recherche Scientifique - Laboratoire d'informatique de l'École Polytec
 hnique (LIX)); and Thomas Daniel (Kinetix)\n---------------------\nSofiBud
 dy: A Soft Mobile Interface for On-Body Interaction\n\nWe present SofiBudd
 y, a soft mobile interface with a lightweight transition module and  a cur
 vature-based inflatable actuator, offering gentle conforming motion with i
 nteractive potential for companionship, assistance, and VR engagement.\n\n
 \nYixuan Li (Tsinghua University, The Hong Kong University of Science and 
 Technology (Guangzhou)); Jie Yu (The Hong Kong University of Science and T
 echnology (Guangzhou)); Chenwan Halley Zhong (The Hong Kong University of 
 Science and Technology (Guangzhou), Northwestern University); Zhaowen Deng
  (The Hong Kong University of Science and Technology (Guangzhou), Dartmout
 h College); Teng Han (Institute of Software, Chinese Academy of Sciences);
  Yi Cai (The Hong Kong University of Science and Technology (Guangzhou)); 
 and Xin Tong (The Hong Kong University of Science and Technology (Guangzho
 u); The Hong Kong university of science and technology, Hong Kong)\n------
 ---------------\nTwo-Stage Vector Sketch Generation Using Structure-to-Det
 ail Optimization\n\nWe proposed a two-stage vector sketch generation frame
 work. \nThe structure stage captures global composition with coarse stroke
 s, while the detail stage  to refine fine-grained details.\n\n\nYuan Ma (J
 apan Advanced Institute of Science and Technology (JAIST)) and Haoran Xie 
 (Japan Advanced Institute of Science and Technology (JAIST), Waseda Univer
 sity)\n---------------------\nCamSketch: Authoring Camera Trajectories for
  3D Virtual Concerts via Example-Derived Codes and User Sketching\n\nCamSk
 etch learns structured camera motion codes from concert footage and enable
 s users to sketch and refine camera paths via an intuitive, scene-aligned 
 interface for immersive 3D trajectories in virtual concerts.\n\n\nWen-Li W
 ei, Yu-Hua Hu, and Jen-Chun Lin (Academia Sinica)\n---------------------\n
 Foveation Improves Payload Capacity in Steganography\n\nOur work investiga
 tes the efficacy and efficiency of utilizing foveated rendering in the dom
 ains of stenography, showing increases in payload capacity and image quali
 ty.\n\n\nLifeng Qiu Lin (University College London), Henry Kam and Qi Sun 
 (New York University), and Kaan Akşit (University College London)\n-------
 --------------\nMourning Skins: Using Generative AI for Speculative Visual
 ization on Climate Change\n\nMourning Skins is a speculative visualization
  using generative AI to explore climate change through imagined species fu
 r transformations presented as time lapse videos\n\n\nYuanjin Zhao and Yec
 hun Peng (Tongji University, Intelligent Big Data Visualization Lab)\n----
 -----------------\nHoverCanvas: Aerial Display Based on a Flying Screen an
 d 100 m Long Range Dynamic Projection\n\nHoverCanvas is a drone-based aeri
 al display that achieves stable projection onto a 2.3m screen from the pro
 jection system 100m away. Its novel use of polar IR markers and high-speed
  mirror control overcomes critical challenges of self-occlusion and projec
 tion latency.\n\n\nShimpei Yamamoto and Hiromasa Oku (Gunma University)\n-
 --------------------\nThrough Their Eyes: A First-Person VR Experience of 
 Alzheimer's Disease\n\nThis study presents an immersive VR simulation syst
 em enabling first-person experience of Alzheimer's daily challenges, and d
 emonstrates its effectiveness in enhancing users' empathy and understandin
 g through mixed-methods evaluation.\n\n\nBaicheng Du, Yuxing Zhang, Boyuan
  Shi, Wantian Gu, Mengyao Guo, and Guanyou Li (Harbin Institute of Technol
 ogy,Shenzhen); Pei Zhang (Shenzhen Institute of Advanced Technology); and 
 Zhiwei Deng (Configreality Technology Co. Ltd.)\n---------------------\nFa
 st and Stable Elastic Dynamics with Large-scale Contact Simulation\n\nTo o
 vercome rich-contact problems, we propose a lightweight approximation sche
 me that significantly reduces the density of the contact system in large-s
 cale elastic simulation.\n\n\nZiqiu Zeng and Siyuan Luo (National Universi
 ty of Singapore), Eric Yang (Gradient), and Fan Shi (National University o
 f Singapore)\n---------------------\nWhispering Shells: An Immersive Inter
 vention for Coastal Conservation\n\nWhispering Shells is an immersive inte
 ractive installation that transforms scientific data about shell depletion
  into a sensory experience. Using Arduino-driven sculpture and real-time p
 rojections, it visualizes human impact on coastal ecosystems and promotes 
 environmental awareness. By merging art,...\n\n\nChaolu Wang, Chujun Zhao,
  Yujia Hu, Yufei Fan, and Yangyang Zheng (Royal College of Art)\n---------
 ------------\nMLLM-guided Training-free Spherical Panorama Generation from
  a Single Image\n\nA training-free pipeline for generating a spherical pan
 orama scene from a single image.\n\n\nYuki Katayama, Ren Togo, Keisuke Mae
 da, Takahiro Ogawa, and Miki Haseyama (Hokkaido University)\n-------------
 --------\nMany-Inbetweens Browsing: Exploring Pre-Generated  Inbetweens fo
 r 2D Animation\n\nThis paper presents a system for exploring pre-generated
  inbetweens via motion trajectories, enhancing efficiency and user experie
 nce in 2D animation production through interactive browsing.\n\n\nJotaro S
 akamiya (The University of Tokyo); Akinobu Maejima (OLM Digital, Inc.; IMA
 GICA GROUP Inc); and Yuki Koyama, I-Chao Shen, and Takeo Igarashi (Univers
 ity of Tokyo)\n---------------------\nVisualizing the Immune Mechanisms: I
 nteractive Storytelling for Pediatric Allergy Engagement\n\nWe developed a
 n interactive storytelling system with tangible media, drawing on insights
  from pediatric healthcare professionals, to translate immune mechanisms a
 nd sublingual immunotherapy (SLIT) into playful, child-centered experience
 s that foster understanding and emotional engagement in long-ter...\n\n\nM
 iki Monzen (Creative Media Research Center, Ritsumeikan University) and Sh
 igenori Mochizuki and Toshikazu Ohshima (College of Image Arts and Science
 s, Ritsumeikan University)\n---------------------\nWhen Stars Shine throug
 h Silence: Affective Visualization as Empowerment for Victims of Subway Ha
 rassment\n\nWhen Stars Shine through Silence is an interactive visualizati
 on using star metaphors to show victims’ courage, online solidarity, and i
 nstitutional response, fostering empowerment and collective awareness.\n\n
 \nYutong Jiang, Xiaoke Yan, Chuyi Zheng, Yang Shi, Qing Chen, and Nan Cao 
 (Tongji University)\n---------------------\nVR K-Heritage Assembler: Explo
 ring Korean Wooden Architecture Through Virtual Gong-po Assembly\n\nVR K-H
 eritage Assembler enables immersive assembly of Gong-po bracket structures
  from twelve Korean national treasure buildings, offering interactive, rep
 eatable, and cost-effective virtual learning to enhance architectural heri
 tage education and understanding.\n\n\nJuwon Lee, Da-un Jung, and Seungjae
  Lee (Electronics and Telecommunications Research Institute (ETRI))\n-----
 ----------------\nAniME: Adaptive Multi-Agent Planning for Long Animation 
 Generation\n\nAniME is a multi-agent system for automatic long-form anime 
 production. The director agent keeps a global memory, and coordinates down
 stream specialized agents which adaptively select tool models for diverse 
 sub-tasks.\n\n\nLisai Zhang, Baohan Xu, Siqian Yang, Mingyu Yin, Jing Liu,
  Chao Xu, Siqi Wang, Yidi Wu, Yuxin Hong, Zihao Zhang, Yanzhang Liang, and
  Yudong Jiang (Bilibili)\n---------------------\nBetween Us: An Interactiv
 e Installation for Parent-Child Relationship Visualization Using NFC and G
 enerative Algorithms\n\nThis installation visualizes parent-child emotiona
 l boundaries by transforming NFC-embedded tokens into generative canopy fo
 rms, merging symbolic visualization, computational aesthetics, and cultura
 l reflection to explore relational dynamics in childhood development.\n\n\
 nNiannian Zhou and Zihan Gao (Communication University of China)\n--------
 -------------\nSketch-in-Scene: Coherent 3D Scene Asset Generation with Sk
 etches\n\nSketch-in-Scene is a novel sketch- and environment-based authori
 ng system that generates 3D assets that are coherent with the surroundings
 .\n\n\nFang Du and Zackary P. T. Sin (Hong Kong Polytechnic University)\n-
 --------------------\nMaterials for the Moon: Developing Physically Based 
 Lunar Regolith Assets to Support Space Training and Simulation\n\nWe prese
 nt work toward a free library of high-fidelity, physically based lunar sur
 face materials to support training and simulation in support of future hum
 an and robotic missions to the Moon.\n\n\nMaximiliano Wardle (University o
 f York, European Space Agency)\n---------------------\nAI-Powered Synthesi
 s of 2.5D Traditional Chinese Paintings\n\nThe method takes a Chinese pain
 ting image and a prompt and generate 2.5D animation that can be viewed fro
 m various views. This approach avoids manual annotation and time-consuming
  editing.\n\n\nXin-Wei Lin, HuiGuang Huang, Dong-Yi Wu, Pin-Chieh Lu, and 
 Sheng-Yi Yao (National Cheng-Kung University); Yu Cao (Saint Francis Unive
 rsity); and Tong-Yee Lee (National Cheng-Kung University)\n---------------
 ------\nColor-Corrected Full Ray-Based Computer-Generated Holography\n\nWe
  propose color-corrected ray-based hologram generation that calibrates col
 or using reference planes and adjusts amplitude with correction factors, a
 chieving a consistent color to a rendered image with precise occlusion.\n\
 n\nJaehong Lee, Youngwoo Kim, and Duksu Kim (Korea University of Technolog
 y & Education (KOREATECH))\n---------------------\nA Framework for Archite
 ctural Geometry Synthesis Using Cellular Automata and Conditioned Diffusio
 n Models\n\nWe present an AI framework that procedurally generates unique 
 3D architectural models and uses diffusion models to create photorealistic
  renderings, enabling rapid, large-scale exploration of novel conceptual d
 esigns.\n\n\nSabri Gokmen and Andrei Vince (University of North Carolina a
 t Charlotte) and Aidan Oh (University of North Carolina at Chapel Hill (UN
 C))\n---------------------\nA Drifting-Object-Based Water Surface Display\
 n\nThis study presents a new example of visual expression on water surface
 s, realized by controlling the positions of drifting objects—such as aquat
 ic plants and fallen leaves—through bubble-induced flow. As a proof-of-con
 cept prototype, we built a 30 cm square water tank equipped with sixt...\n
 \n\nKanato Takahashi, Kengo Imai, Reiji Nakano, Taiyo Tamura, Tsubasa Kita
 jima, and Yuichiro Katsumoto (Tokyo Denki University)\n-------------------
 --\nSimulating 3D Thermal Fluid Dynamics in Data Centers with Soft-Constra
 ined Physics-Informed Graph Neural Network\n\nData Driven Physics informed
  learning for Temperature field transient simulation in hot aisle region o
 f a data center.\n\n\nyidi Wang (NVIDIA, Singapore Institute of Technology
 ); Aik Beng Ng and Simon See (NVIDIA); and Daniel Wang and Frank Guan (Sin
 gapore Institute of Technology)\n---------------------\nThe Stage Comes to
  You: A Real-Time Tele-Immersive System with 3D Point Clouds and Vibrotact
 ile Feedback\n\nWe present a real-time tele-immersive system that transmit
 s 3D point clouds and performers’ footstep vibrations to remote audiences 
 with less than 100 ms latency. Combining synchronized LiDAR–camera capture
 , lightweight flicker reduction, and view-dependent rendering, the system 
 streams ...\n\n\nTakahiro Matsumoto, Takahiro Kusabuka, Hiroshi Chigira, K
 azuhiko Murasaki, Kakagu Komazaki, Masafumi Suzuki, and Masakatsu Aoki (NT
 T)\n---------------------\nPoetry Space: Visualization of Poetry Recitatio
 n Emotions in VR\n\nStudying emotion visualization for VR poetry recitatio
 n training, we designed 'Poetry Space'. This system captures the reciter’s
  emotions and voice real-time, transforming them into a virtual mountain l
 andscape.\n\n\nXuanmiao Zhang (School of New Media Arts & Design, Beihang 
 University) and Hang Yu, Jie Chen, Zhixiang Fu, and Yong Hu (Beihang Unive
 rsity)\n---------------------\nGenerative Tonal Art: An Artistic Rendering
  Workflow using Programmable TAMs\n\nThis work introduces Generative Tonal
  Art Maps (GTAMs), a workflow that bridges generative art and non-photorea
 listic rendering, enabling artists to design customizable textures in p5.j
 s and apply them for real-time stylized rendering in Blender.\n\n\nYun-Che
 n Lee (National Yang Ming Chiao Tung University) and June-Hao Hou (Graduat
 e Institute of Architecture, National Yang Ming Chiao Tung University)\n--
 -------------------\nLayerPack: Training-Free Animation Inbetweening with 
 Cross-Layer Consistency\n\nWe propose "LayerPack", a method for training-f
 ree layer-wise frame interpolation.\n\n\nAtsushi Maruyama (The University 
 of Tokyo); Sosui Koga (Nara Institute of Science and Technology); Akinobu 
 Maejima (OLM Digital, Inc.; IMAGICA GROUP Inc.); and Yuki Koyama, I-Chao S
 hen, and Takeo Igarashi (The University of Tokyo)\n---------------------\n
 When Birds Bow: An Interactive Audio-Visual Installation for Suona Culture
  via Collaborative Tangible User Interface\n\nAn interactive audio-visual 
 installation that reimagines Suona culture through collaborative tangible 
 interaction, transforming ritual performance into a symbolic journey of bi
 rd calls and phoenix rebirth.\n\n\nYuyao Heng, Yi Zhang, and Zihan Gao (Sc
 hool of Animation and Digital Arts,   Communication University of China)\n
 ---------------------\nTowards Wearable-Free Exergame for Orthostatic Dysr
 egulation: Improving Motion Detection with Smartphone-Only Calibration\n\n
 This study develops a smartphone-only exergame for adolescents with orthos
 tatic dysregulation to reduce sensor burden while maintaining detection ac
 curacy, and compared it with a conventional method using an external senso
 r.\n\n\nHitomi Miyazaki, Saki Sakaguchi, Tetsuaki Baba, and Kumiko Kushiya
 ma (Tokyo Metropolitan University)\n---------------------\nDifferentiable 
 Rendering Framework for Retroreflective Model Construction in Mid-Air Imag
 e Simulation\n\nWe propose a framework for constructing retroreflective mo
 dels for mid-air image simulation. These models are derived from photograp
 hs of mid-air images and optimized using differentiable rendering. They re
 produce viewing-dependent luminance and blur more accurately than conventi
 onal models, reducin...\n\n\nRyo Sugawara and Naoya Koizumi (The Universit
 y of Electro-Communications)\n---------------------\nEnabling Visually Awa
 re Conversational Data Visualization Through LLM Augmentation\n\nWe introd
 uce a method that makes Large Language Models both data-aware and visually
 -aware by augmenting them with structured metadata. Visual features and te
 xtual descriptions are combined into a compact JSON file, enabling accurat
 e, context-specific answers without fine-tuning. Tested on geospatial...\n
 \n\nOmar Mena (King Abdullah University of Science and Technology (KAUST))
 , Alexandre Kouyoumdjian (KAUST), Lonni Besançon (Linköping University), M
 ichael Gleicher (University of Wisconsin Madison), Ivan Viola (King Abdull
 ah University of Science and Technology (KAUST)), and Anders Ynnerman (Lin
 köping University)\n---------------------\nHAVM: Designing 3D Hybrid-Aniso
 tropic-Voronoi Cells for Microstructured Metamaterial\n\nWe introduce a no
 vel computational framework for designing hybrid-anisotropic-Voronoi micro
 structures governed by a set of fully geometric-driven parameters, enablin
 g comprehensive control for 3D cell shape, orientation, density, hybridity
 , and thickness.\n\n\nHaikuan Zhu, Sikai Zhong, Jing Hua, and Jimmy Ching-
 Ming Chen (Wayne State University); Yongjie Jessica Zhang (Carnegie Mellon
  University); Charlie C.L. Wang (The University of Manchester); and Zichun
  Zhong (Wayne State University)\n---------------------\nGPStroke: A Walkin
 g Promotion Method using GPS Art with Multiple Walking Paths\n\nGPStroke e
 nables people to create GPS art by collecting fragmented daily walking pat
 hs as "strokes" and later combining them into artworks, promoting walking 
 without requiring dedicated exercise time.\n\n\nEtsuki Omoto and Koji Tsuk
 ada (Future University Hakodate)\n---------------------\nImproving Localiz
 ation with Localization-guided Structure-from-Motion\n\nOur method, LGSfM,
  is the first approach to optimize SfM-based 3D models for visual localiza
 tion. Experiments show LGSfM achieves consistent improvements in real-worl
 d localization across both indoor and outdoor datasets.\n\n\nYu-Yi Lin, Ka
 i-Li Wang, and Kuan-Wen Chen (National Yang Ming Chiao Tung University)\n-
 --------------------\nTentacles Fever: Sound-Responsive SMA Actuators\n\nT
 entacles Fever is an installation that visualizes sound and music through 
 the movement of tentacle-like actuators made with shape-memory alloy (SMA)
 . Each actuator bends smoothly in multiple directions, synchronized with m
 usical rhythms, producing an organic and performative expression—as if ...
 \n\n\nAkira Nakayasu (Tokyo Metropolitan University)\n--------------------
 -\nTunTun Diary: Exploring AI-Generated Storytelling and Virtual Companion
 ship for Nightmare Relief\n\nTunTun Diary is a playful mobile game where a
 n alien puppy transforms nightmares into comic stories, offering comfort a
 nd emotional relief through AI storytelling and virtual companionship.\n\n
 \nZixin Wang (University of Michigan); Shan Luo, Siyu Tian, and Yunxi Cai 
 (China Academy of Art); Long Ling (Tongji University); Muhan Xu (Universit
 y of the Arts London); Jiaxing Yu and Guanting Lu (Zhejiang University); J
 iajing Zhu (China Academy of Art); and Liping Luo (Zhejiang Sci-Tech Unive
 rsity)\n---------------------\nTopology-Aware Polymorphism for Embedded Sh
 ading Languages\n\nTo solve the scalability and hot-reloading challenges o
 f instance polymorphism in ESLs, we've developed a topology-aware polymorp
 hic architecture that drastically cuts compilation and rendering times.\n\
 n\nLing Zhu (Unity), Liulei Zou (Jiangsu University of Technology), Shaoku
 n Zheng (Huawei), and Kun Xu (Tsinghua University)\n---------------------\
 nProjecting Softness: Finger-Bending Illusions in Tangible Projection Mapp
 ing\n\nWe propose a projection technique that induces finger-bending illus
 ions using luminance-based motion cues, enhancing perceived softness on ri
 gid objects without physical deformation or wearables in augmented reality
  contexts.\n\n\nNaoki Hashimoto and Shota Yamagishi (The University of Ele
 ctro-Communications)\n---------------------\nNovel active force feedback g
 love and control algorithm for a realistic virtual handshake\n\nThis study
  presents the active force feedback glove and its control algorithm, succe
 ssfully guiding users' fingers into a handshake posture, allowing users to
  experience realistic virtual handshakes in XR environments.\n\n\nDaegeun 
 Park, Juyoung Kim, Yeongjae Choi, and Sung-Uk Jung (Electronics and Teleco
 mmunications Research Institute (ETRI))\n---------------------\nDesign of 
 Nonverbal Symbolic Visual Cues for VR Social Interaction Based on AI Seman
 tic Recognition\n\nThis project presents an AI-driven nonverbal visual fee
 dback system for VR social interaction, enhancing emotion, clarity, and im
 mersion, with validated user improvements and future scalability across di
 verse scenarios.\n\n\nYulu Lu (School of New Media Art and Design, Beihang
  University, Beijing, China,); Zhikai Wen (State Key Laboratory of Virtual
  Reality Technology and Systems, Beihang University); Xinyu Su (School of 
 Humanities and Social Sciences ,Beihang University); and Yong Hu (School o
 f New Media Art and Design, Beihang University, Beijing, China,; State Key
  Laboratory of Virtual Reality Technology and Systems, Beihang University)
 \n---------------------\nCan Vision-Language Models Perceive "Lookalike Ef
 fects" Between Humans and Cartoons as We Do?\n\nDiscover how Vision-Langua
 ge Models and humans see differently when matching cartoon avatars to real
  people — revealing the gaps between machine and human vision.\n\n\nChenzh
 uo Xiang, Minghui Liu, Qi Xin, Jihong Jeung, and Qianyao Xu (Tsinghua Univ
 ersity)\n---------------------\nPaperMythXR: Embodied Mixed-Reality Storyt
 elling via Chinese Paper-Cutting\n\nPaperMythXR fuses real paper-cutting w
 ith mixed reality to drive Pangu's myth. A Read-Create-Advance loop and hu
 man-AI production pipeline unite book, tools, and MR, boosting immersion, 
 cultural connection, and creative agency.\n\n\nShantong Zhu, Wenjuan Chen,
  Jiayi Wu, Xingguang Mi, Wenxuan Kou, Yilin Zhang, and Congyun Tang (Commu
 nication University of China)\n---------------------\nCapturing Size-Aware
  Draping via Per-Garment 2D Try-On\n\nWe propose a size-aware virtual try-
 on approach that simulates the degree of tight or loose draping on specifi
 c body sizes directly in 2D, enabling efficient interactive applications. 
 Our per-garment method follows a capture-and-synthesis pipeline: we first 
 capture garment appearances from video und...\n\n\nJingyuan Liu, Zaiqiang 
 Wu, Yechen Li, and Takeo Igarashi (The University of Tokyo)\n-------------
 --------\nSchool Legacy: Reconstructing Memory Through AI-augmented Social
  XR\n\nRelive memories in VR. Scan, transform, connect.\n\nCreate 3D time 
 capsules, embody your avatar, and share stories in a nostalgic, shared vir
 tual world.\n\nPast meets present. Memory becomes magic.\n\n\nYangli Liu (
 Seelove Technology, Northeastern University); Mehdi Akhondi (Islamic Azad 
 University); Eraulia Nadinda Rachmat (Bandung Institute of Technology); Me
 ngyao Guo (Harbin Institute of Technology, Shenzhen); Ibrahim Olalekan Jim
 oh (University Of Ilorin); and Haowei Xiong (University of the Arts London
 )\n---------------------\nAR-Supported Online Piano Lessons with Bidirecti
 onal Visualization of Teacher and Student Hand Movements\n\nWe propose a b
 idirectional AR-supported online piano lesson system where teachers and st
 udents share hand movements in real time. Our preliminary prototype demons
 trates potential for remote education and interactive learning.\n\n\nMayuk
 a Yamashita and Soh Masuko (Shibaura Institute of Technology)\n-----------
 ----------\nLiving Literature: A Collaborative VR Experience for Enhancing
  Shared Reading Through Affective and Generative Feedback\n\n“Living Liter
 ature” revolutionizes e-book reading by blending VR, haptics, audio, and b
 iosensing into a shared, emotion-responsive experience, enhancing engageme
 nt, immersion, and discussion depth in collaborative reading clubs\n\n\nHa
 itian Liu (University of Auckland, School of Computer Science); Tamil Selv
 an Gunasekaran (University of Auckland, Empathic Computing Lab); Mark Bill
 inghurst (University of Auckland, University of South Australia); and Yun 
 Suen Pai (University of Auckland, School of Computer Science)\n-----------
 ----------\nA Bio-sensing System for Visualizing Material Perceptions thro
 ugh Physiological Computing and Geometric Encoding\n\nWe propose a biosens
 ing system that converts material perception into visual and geometric enc
 odings. This system sets grounds for future scientific information visuali
 zation, connecting human perception and tangible material expression.\n\n\
 nMirna Zordan, Enza Migliore, Minghong Zhang, and Peiyue Lin (Southern Uni
 versity of Science and Technology)\n---------------------\nETCE: Efficient
  Two-Stage Concept Erasure for Text-to-Image Diffusion Models\n\nOur paper
  proposes ETCE, a novel concept erasure algorithm for text-to-image models
 , designed to prevent the generation of illegal or copyright-infringing co
 ntent.\n\n\nQi-Ang Hu (Tsinghua University); Nisha Huang (Tsinghua Univers
 ity, Pengcheng Laboratory); Yizhou Lin (Tsinghua University); Jie Guo (Pen
 gcheng Laboratory); Xiu Li (Tsinghua University); and Tong-Yee Lee (Nation
 al Cheng-Kung University)\n---------------------\nInvertible Video Signal 
 Processing\n\nWe make video signal processing invertible, for easily acces
 sible raw data.\n\n\nYazhou Xing, Zian Qian, Ji Zhou, and Qifeng Chen (HKU
 ST)\n---------------------\nFast First-Hit Determination for Large and Com
 posite Sparse Volumetric Data Structures with Mesh Shaders\n\nTask–mesh sh
 ader pipeline maps sparse volume for first-hit rendering; L1 acceleration-
 structure jumps and tile-local HDDA remove intermediary geometry buffers, 
 delivering 1.5x average speedups over vertex-shader baselines while preser
 ving G-buffer equivalence.\n\n\nMehmet Oguz Derin (Morgenrot) and Takahiro
  Harada (Advanced Micro Devices, Inc. (AMD); Morgenrot)\n-----------------
 ----\nVideo2Song: Video-to-Song Generation via Retrieval-Augmented Prompt 
 Generation\n\nAn end‑to‑end framework, Video2Song, for video soundtrack ge
 neration that combines a VLM and retrieval‑augmented generation to extract
  semantic, stylistic, and rhythmic cues, conditioning a LLM to guide music
  synthesis.\n\n\nZijiao Yin (School of Artificial Intelligence, University
  of Chinese Academy of Sciences; MAIS, Institute of Automation, Chinese Ac
 ademy of Sciences) and Sifei Li and Weiming Dong (MAIS, Institute of Autom
 ation, Chinese Academy of Sciences; School of Artificial Intelligence, Uni
 versity of Chinese Academy of Sciences)\n---------------------\n“I’m Alway
 s Here”: An Affective Companion Agent for Novice Drivers\n\nThis research 
 introduces an affective in-vehicle companion agent that proactively provid
 es situational guidance and emotional support for novice drivers through r
 eal-time emotion recognition, anthropomorphic animation, and voice interac
 tion, demonstrating enhanced user experience and satisfaction in...\n\n\nW
 ei GONG and Stephen Jia Wang (School of Design, The Hong Kong Polytechnic 
 University)\n---------------------\nEmotion-Aware Autonomous Driving:  A V
 R-Based Multimodal Analysis Using Gaze and Physiological Signals\n\nThis s
 tudy uses VR-based autonomous driving scenarios with Calm and Tense styles
  to analyze multimodal responses. A Random Forest classifier achieved 97.1
 % accuracy, demonstrating feasibility of emotion-aware driving systems.\n\
 n\nMI CHANG, JIWOO HAN, WOOJIN KIM, and DAESUB YOON (Electronics and Telec
 ommunications Research Institute (ETRI))\n---------------------\nA Card-Ba
 sed Design Toolkit for Co-Designing Remote Multimodal Communication and In
 teraction Systems with Older Adults and Experts.\n\nThis study presents a 
 tangible card-based toolkit enabling older adults to co-design XR spatial–
 physical interactions, translating lived experiences into actionable conce
 pts that enhance emotional well-being and long-distance social connectedne
 ss.\n\n\nDezijian Zhou (Design Lab, Sydney School of Architecture, Design 
 and Planning, The University of Sydney); Yuanyuan Deng (University College
  London (UCL)); Yi Zhao (Design Lab, Sydney School of Architecture, Design
  and Planning, The University of Sydney); Jialing Xie (Sydney School of Ar
 chitecture, Design and Planning, The University of Sydney); Zihan Peng and
  Wanli Guo (University College London (UCL)); and Anastasia Globa (Sydney 
 School of Architecture, Design and Planning, The University of Sydney, Syd
 ney, NSWDesign Lab, Sydney School of Architecture, Design and Planning, Th
 e University of Sydney)\n---------------------\nZenith: Procedural and Dif
 fusion-Driven Generation of Multi-Layered Top-Down Maps\n\nZenith is a pip
 eline that automates the creation of multi-layered top-down maps for World
  of Warcraft dungeons. It combines procedural geometry processing with fin
 e-tuned diffusion models to generate artistic layers (line art, shadows, h
 ighlights). A novel multi-encoder ControlNet ensures spatial co...\n\n\nEh
 san Sayyad, Trevor Rothman, and Zhen Zhai (Blizzard Entertainment)\n------
 ---------------\nDeanthropocentric Interfaces: Plant-Driven Interaction De
 sign in the Metaverse Era\n\nBCTI framework transforms plants into autonom
 ous generative agents within the metaverse, driving a bio-renaissance wher
 e ecological processes create art, demonstrating 150% growth in plant-driv
 en digital creativity.\n\n\nZe Gao (Cyanpuppets); Yuan Zhuang (Shandong Un
 iversity); Mengyao Guo (Harbin Institute of Technology, Shenzhen); and Kun
 qi Wang (Jinan Foreign Language School)\n---------------------\nFloorplan 
 Generation with Graph Beta Diffusion\n\nOur single-stage Graph Beta Diffus
 ion generates residential floorplans as graphs, jointly modeling coordinat
 es, room types, and edges. An unsupervised Manhattan-alignment loss improv
 es FID over GAN baselines with simpler, end-to-end pipeline.\n\n\nYusuke T
 akeuchi (Tetraz, Inc.; University of Tokyo) and Qi An and Atsushi Yamashit
 a (University of Tokyo)\n---------------------\nA Practical Encoding Appro
 ach for Texture Compression: Combining Multi-Processing and Multi-Threadin
 g\n\nA pratical parallel encoding method combining multi-processing and mu
 lti-threading, with size-based file sorting to reduce bottlenecks and impr
 ove efficiency in large-scale texture compression.\n\n\nHyeon-ki Lee and J
 ae-Ho Nah (Sangmyung University)\n---------------------\nEdible Chromatic 
 Caustics: Designing Colored Caustics of Jelly via Differentiable Rendering
 \n\nThis study proposes a method for designing and fabricating jellies wit
 h desired color caustic patterns by embedding colored jelly inclusions in 
 a transparent base jelly.\n\n\nDaiki Inokoshi and Takashi Ijiri (Shibaura 
 Institute of Technology)\n---------------------\nMindEye: A VR Imagery Tra
 ining System for Individuals with Acquired Visual Imagery Impairments and 
 Mild Aphantasia\n\nMindEye is a VR-based therapeutic system combining guid
 ed imagery and spatial-semantic association to restore visual imagery abil
 ities in individuals with aphantasia and acquired imagery impairments.\n\n
 \nTian Wang, Chenyang He, Ruici Song, Jinghan Zhang, and Yijing Xu (Harbin
  Institute of Technology, Shenzhen); Pei Zhang (Shenzhen Polytechnic Unive
 rsity); Mengyao Guo and Guanyou Li (Harbin Institute of Technology, Shenzh
 en); and Zhiwei Deng (Configreality Technology Co.Ltd.)\n-----------------
 ----\nImage-Similarity-Based Scale Estimation for 3D Model\n\nAuto 3D obje
 ct scaling is important and difficult. We propose an image-similarity-base
 d estimation method, delivering faster and more precise model resizing for
  AI model and scene generation.\n\n\nTianrui Hu (Beijing University of Pos
 ts and Telecommunications; School of Creative Media, City University of Ho
 ng Kong); Jiaao Yu (Beijing University of Posts and Telecommunications, Te
 ncent Hunyuan); and Wenjun Hou (Beijing University of Posts and Telecommun
 ications)\n---------------------\nCreative Hair Style Generation and Editi
 ng\n\nDiffusion-based system for interactive hairstyle design, enabling hy
 brid style blending, realistic recoloring, and drag-based volumetric edits
  for creative, controllable, and photorealistic virtual hairstyling applic
 ations.\n\n\nSheng-Yi Yao, Dong-Yi Wu, Qian-Hui Lin, Yu-Hsuan Kung, Ci-Hua
 n Huang, HuiGuang Huang, and Xin-Wei Lin (National Cheng Kung University);
  Yu Cao (Saint Francis University); and Tong-Yee Lee (National Cheng Kung 
 University)\n---------------------\nSimulating Loneliness: Human–Generativ
 e Agent Comparative Insights\n\nGenerative agents simulate loneliness, mat
 ching human data on standardized psychological scales, presenting a comput
 ational approach to model subjective and dynamic psychological experiences
 .\n\n\nHiroki Fujiwara (Advanced Institute of Industrial Technology, Sieme
 ns K.K) and Daisuke Kasai, Toshiya Goto, and Ken-ichi Tabei (Advanced Inst
 itute of Industrial Technology)\n---------------------\nAcoustic Form: Enh
 ancing the Design Process for Assistive Earwear Jewelry via Material-Centr
 ic Computational Sound Analysis\n\nWe present a simulation-based design wo
 rkflow for assistive earwear jewelry. The Green Voice ear cuff demonstrate
 s how material properties shape passive sound amplification, integrating a
 coustic analysis with aesthetics.\n\n\nArefeh Ahmadi and Emil Polyak (Drex
 el University)\n---------------------\nAMV3D: Adaptive Multi-Layer Volumet
 ric 3D for DICOM\n\nAMV3D is a lightweight volumetric format/renderer for 
 medical CT/MRI that creates a multi-resolution single-file.On PC/XR headse
 ts it loads instantly, reduces memory, and improves timeframe over 3D text
 ures and sparse volumes.\n\n\nLeonardo Locatelli (Adapta Studio); Matteo V
 aloriani (FifthIngenium, Politecnico di Milano); Manuel Pezzera (FifthInge
 nium); Simona Perotto (Politecnico di Milano, ADAPTA studio); and Francesc
 o Clasadonte (FifthIngenium)\n---------------------\nEchoes of the Matchlo
 ck: Playing Shooting Games Using Cultural Heritage in Mixed Reality\n\nEch
 oes of the Matchlock is a Mixed Reality system making Japanese matchlock f
 irearms engaging for younger audiences through gamified education, bridgin
 g the gap in cultural heritage transmission.\n\n\nWenze Song and Takefumi 
 Hayashi (Kansai University)\n---------------------\nThe Effect of Shape Di
 fferences on Object Impressions and Their Relation to Weight Perception\n\
 nThis study explores how object shape impressions affect weight perception
  in VR, aiming to clarify the relationship and underlying factors between 
 shape and perceived weight.\n\n\nKosei Tomioka (Utsunomiya University), Na
 oki Hashimoto (The University of Electro-Communications), and Mie Sato (Ut
 sunomiya University)\n---------------------\nWhere Things Become Imageries
 ：Visualization Experience of Chinese Poetic Imageries in Immersive VR\n\nT
 he project visualizes Tang poetry imagery through a VR interactive system,
  enabling immersive exploration of poetic imagery.\n\n\nHuantong Xian, Lik
 un Zhao, and Xiaohe Peng (School of New Media Art and Design,Beihang Unive
 rsity) and Shuo Yan (School of New Media Art and Design,Beihang University
 ; State Key Laboratory of Virtual Reality Technology and Systems, Beihang 
 University)\n---------------------\nTextFX: Reconstructing Editable Styliz
 ed Typography Through Parametric Text Effect Extraction\n\nTextFX introduc
 es a structured pipeline for stylized text reconstruction, reducing design
  effort while achieving strong performance. User feedback validate effecti
 veness, with future work targeting learnable parameters and prompt-driven 
 creative typography.\n\n\nVeeramanohar Avudaiappan (Department of Electric
 al and Electronics Engineering, Amrita School of Engineering, Coimbatore, 
 Amrita Vishwa Vidyapeetham); Alooysius Vitalian J (Department of Cyber Sec
 urity, Amrita School of Engineering, Coimbatore, Amrita Vishwa Vidyapeetha
 m); and Ritwik Murali (Department of Computer Science and Engineering, Amr
 ita School of Computing, Coimbatore, Amrita Vishwa Vidyapeetham)\n--------
 -------------\nBridging Visual Rhetoric and Computer Vision: A Large-Scale
  Analysis of the Design of Japanese and Original-Language Film Posters\n\n
 This study bridges visual rhetoric and computer vision, presenting the fir
 st large-scale comparative analysis of Japanese and original-language film
  posters, revealing cultural differences in design through quantitative im
 age analysis.\n\n\nNano Kojima and Xinru Zhu (Institute of Science Tokyo)\
 n---------------------\nAeroVis3R: Geometry-Aware Vision–Language Models f
 or 3D Reasoning over UAV Landmark Videos\n\nAeroVis3R introduces geometry-
 aware vision–language models for UAV landmark videos, uniting 3D reconstru
 ction and GPT reasoning. We also release a DJI Mini 3 Pro landmark dataset
  with Wikipedia annotations.\n\n\nCewen Deng (HKUST)\n--------------------
 -\nHands in Bloom: An Auditory-Accessible Immersive Experience Based on Ch
 inese Sign Language Recognition and Gesture Tracking\n\nThis project integ
 rates Chinese Sign Language into immersive interaction, enhancing expressi
 veness and inclusivity while enabling barrier-free communication that brid
 ges Deaf and hearing communities beyond natural gestures and instruction.\
 n\n\nZhu Zhu (School of New Media Art and Design, Beihang University; Beih
 ang University) and Shuo Yan (School of New Media Art and Design, Beihang 
 University; State Key Laboratory of Virtual Reality Technology and Systems
 , Beihang University)\n---------------------\nPart-Aware Embedding for Fin
 e-Grained 3D Shape Retrieval\n\nWe propose a novel part-aware 3D shape ret
 rieval framework that improves fine-grained search accuracy by embedding b
 oth whole objects and their segmented parts into a shared metric space.\n\
 n\nSuwoong Lee (Electronics and Telecommunications Research Institute (ETR
 I), Korea Advanced Institute of Science and Technology (KAIST)); Da-un Jun
 g, Juwon Lee, and Seungjae Lee (Electronics and Telecommunications Researc
 h Institute (ETRI)); and Junmo Kim (Korea Advanced Institute of Science an
 d Technology (KAIST))\n---------------------\nWeight Illusion from a “1st+
 3rd” Person Perspective: Integrating the Strengths of the 1st and 3rd Pers
 on Perspectives\n\nIn this paper, with a new weight comparison task, we in
 vestigate the effect of viewpoint position on the weight illusion in detai
 l.\n\n\nSakurako Uchiyama (Utsunomiya University), Naoki Hashimoto (The Un
 iversity of Electro-Communications), and Mie Sato (Utsunomiya University)\
 n---------------------\nTactile Data Recording System for Clothing with Mo
 tion-Controlled Robotic Sliding\n\nWe developed a robotic system that coll
 ects non-destructive, motion-labeled tactile data, overcoming prior method
 s that damaged garments and enabling accurate machine learning classificat
 ion.\n\n\nMichikuni Eguchi (Cluster Metaverse Lab, Tsukuba University); Ta
 kekazu Kitagishi (The University of Tokyo, ZOZO Research); Yuichi Hiroi (C
 luster Metaverse Lab); and Takefumi Hiraki (Cluster Metaverse Lab, Tsukuba
  University)\n---------------------\nSee, Sense, Spark: Profiling Granular
 ity in Proactive AI for Creative Research Ideation\n\nWhat if proactive AI
  knew you so precisely that it inspired deeper curiosity instead of limiti
 ng it? In our study with See, Sense, Spark, a smart-glasses-based proactiv
 e AI, we compared coarse and fine-grained user profiling during creative r
 esearch ideation. Surprisingly, participants with fine-gra...\n\n\nGahui K
 im (Sungkyunkwan University, Immersive media engineering); Yebom Choi (Sun
 gkyunkwan University, Department of Human-Artificial Intelligence Interact
 ion); and Changjun Lee (Sungkyunkwan University, School of Convergence)\n-
 --------------------\nImproving Registration-Based High-Resolution Panoram
 a Depth Estimation with Enhanced Fine Details\n\nWe propose a three-stage 
 method improving prior stitching depth generation, introducing SeamRemove 
 for seam removal and SOTA accuracy, and applying existing enhancement tech
 niques to produce visually superior panoramic depth maps.\n\n\nYu Sheng Ya
 ng and Chi Han Peng (National Yang Ming Chiao Tung University)\n----------
 -----------\nAn Abstract Form Guided Method for Image Generation without T
 raining\n\nWe proposed a training-free method for controlling image genera
 tion using abstract forms, such as scribbles and sketches, which allows fo
 r precise control of shape and color compared to text prompts.\n\n\nDanqin
 g Pan and Shun Kaizu (Sony Semiconductor Solutions Corporation)\n---------
 ------------\nA Participant-Driven Spatial Narrative for Cultural Exhibiti
 ons\n\nVisitors co-author stories in a 270-degree immersive room. Instant 
 choices reshape projections into branched endings. This participant-driven
  design sparks curiosity and informs a reusable framework for storytelling
  in cultural exhibitions.\n\n\nBokyin Lo, Kanghong Zhong, Takatoshi Yoshid
 a, Dunya Chen, and Kouta Minamizawa (Keio University Graduate School of Me
 dia Design)\n---------------------\nCustomer-Service Mirrored Avatar That 
 Comes Beside You via Directional Spatial Audio\n\nWe propose an interactiv
 e avatar system with a half-mirror display and parametric loudspeakers. It
  creates a “whisper-beside-you” 3D audio experience, enhancing perceived c
 loseness and immersion for customer service.\n\n\nNanako Matsuda, Hyuma Au
 chi, Tadashi Ebihara, Naoto Wakatsuki, and Keiichi Zempo (University of Ts
 ukuba)\n---------------------\nPredicting Passenger Anxiety with a Spatio-
 Temporal PIGNN  in VR Autonomous Driving\n\nWe propose a spatio-temporal p
 hysics-informed graph neural network (PIGNN) that predicts passenger anxie
 ty using gaze and context data collected in a fully autonomous virtual rea
 lity driving simulation.\n\n\nMI CHANG, JIWOO HAN, WOOJIN KIM, and DAESUB 
 YOON (Electronics and Telecommunications Research Institute (ETRI))\n-----
 ----------------\nAI-Powered Sculpting: Reviving Digital Sculpture with Ge
 nerative Workflows\n\nThis paper introduces a hybrid workflow combining ZB
 rush sculpting with diffusion models, enabling faster stylistic exploratio
 n, reproducible co-creation, and AI-assisted character development while p
 reserving creative authorship.\n\n\nXue Bai, Allan Fowler, and Athirah Zai
 ni (University of Auckland)\n---------------------\nSpatio‑Temporal 3D Gau
 ssian Splatting Alignment via Robust Features and Randomized Sampling\n\nT
 his study proposes a novel method for aligning 3DGS data by leveraging its
  unique characteristics. Our approach uses a sampling and cleaning process
  to efficiently extract key points, enabling high-accuracy and fast regist
 ration without GPU. Experiments show our method outperforms existing techn
 iqu...\n\n\nYuji Okamoto (Chubu Electric Power Co., Inc.); Yuki Oshio, Aka
 ne Shiki, Yuya Fukiharu, Ai Ito, and Shota Imai (UPHASH Inc.); and Takeshi
  Nakamura and Yoshiyuki Numata (Chubu Electric Power Co., Inc.)\n\nRegistr
 ation Category: Enhanced Access, Exhibit & Experience Access, Experience H
 all Exhibitor, Full Access, Full Access Supporter, Trade Exhibitor
END:VEVENT
END:VCALENDAR