XRMemory'25: 2nd International Workshop on Spatial Memory in XR:
Defining and Capturing Memory for Interactive Playback in XR

_________________________________________________________________

IEEE ISMAR 2025 - October 12th (Sunday) 8:30 - 12:30

#103, Daejeon Convention Center (DCC), Daejeon, Korea

Overview

XRMemory'25: 2nd International Workshop on Spatial Memory in XR focuses on the emerging challenge of recording and representing spatial experiences in a way that supports adaptive playback, particularly within XR environments. While photos and videos have long served as passive memory aids, the rise of AR/VR technologies and spatial computing enables the capture of experiences that are not only immersive but also interactive and reconfigurable—supporting interaction with recorded scenes, virtual avatars, AI agents, and other immersed and non-immersed humans.

This workshop aims to bring together experts across AI, XR, graphics, computer vision, and large language models (LLMs) to investigate the technological foundations for capturing, recreating, and sharing these dynamic spatial experiences. By fostering discussions on how these interdisciplinary technologies can enable richer, more vivid forms of memory retention and experience sharing, we hope to reshape the way we remember memories across the boundaries of time and space.

Methods for capturing and structuring dynamic spatial experiences in XR environments.
User perception and cognitive modeling of recorded spatial experiences across time and media.
Systems and frameworks for interactive and adaptive playback of recorded spatial memories.
Techniques for integrating virtual avatars, AI agents, and real users into reconfigurable memory scenes.
Standardization and representation models for storing and sharing immersive spatiotemporal experiences.

Workshop Schedule

XRMemory'25 will be organized as a half-day in-person workshop on October 12th, Morning at IEEE ISMAR 2025.

The tentative schedule of the XRMemory'25 workshop schedule is as follows:

Registration & Introduction (10 min)
Keynote Presentation (20 min) - Sung-Hee Lee, Professor, KAIST
Paper Presentation Session 1: Cognition and Perception in XR (60 min)

Adaptive Virtual Reality Space Design Based on Real-Time User Emotion
Authors: Sun Ah Bae (KAIST), Yujeung Eum (KAIST), Hyun Woo Kim (KAIST), Yeongjo Kim (KAIST), Da Hye Jeon (KAIST), Jae Young Jang (KAIST), Seung Hyun Cha (KAIST)
Effect of Ceiling Height and Spatial Sharing Methods on Cognitive Load During Collaboration in Extended Reality
Authors: Hyemin Shin (KAIST), Sangmin Park (KAIST), Jinwook Kim (KAIST), Jungmin Lee (KAIST), Jeongmi Lee (KAIST)
Map Formats and Landmark Cues: Their Influence on Spatial Memory and Wayfinding in Virtual Reality
Authors: Sabah Boustila (Manchester Metropolitan University), Greg A. Jamieson (University of Toronto), Paul Milgram (University of Toronto)
Distance-Adaptive Visual Guidance for Spatial Awareness Formation in Out-of-View Augmented Reality
Authors: Inyoung Choi (Chonnam National University), Hieyong Jeong (Chonnam National University), Choonsung Shin (Chonnam National University)
Guidelines for combining user preference and interaction experience to visualize the spatial imagination in virtual reality memory training
Authors: Cancan Jin (Zhejiang Wannli University), Yanze Gao ( Zhejiang Wannli University), Zirui Yu (Zhejiang Wannli University), Ningning Xu (Zhejiang Wannli University)

Break (20 min)
Paper Presentation Session 2: AI-Powered Interaction and Memory Augmentation (60 min)

Adaptive Playback of Pointing Gestures for Different Target Positions
Authors: Jiho Kang (KAIST), Sung-Hee Lee (KAIST)
Blankscope: An AI-Powered XR Binocular for Exploring Reality and Memory through Tangible Interaction and Reconstructive Perception
Authors: Chiun Lee (Harvard University), Qingyun Liu (Harvard University), Krystal Montgomery (Massachusetts Institute of Technology)
MemoGlove: Capturing and Recreating XR Memory with Haptic Interaction Traces
Authors: Qingyun Liu (Harvard University), Yuhan Wang (Harvard University)
Few-Shot Action Scene Graph Generation from Video via Multimodal Language Models for Structuring Spatial Experience
Authors: Jinseok Hong (KAIST), Hyerim Park (KAIST), Heejeong Ko (KAIST), Woontack Woo (KAIST)
Designing Memory-Augmented AR Agents for Spatiotemporal Reasoning in Personalized Task Assistance
Authors: Dongwook Choi (Yonsei University), Taeyoon Kwon (Yonsei University), Dongil Yang (Yonsei University), Hyojun Kim (Yonsei University), Jinyoung Yeo (Yonsei University)

Panel Discussion – About Defining and Capturing XRMemory (30 min)
Summary & Closing (10 min)

Keynote Speaker

Sung-Hee Lee

Head at KAIST GSCT

Sung-Hee Lee is the Head of the Graduate School of Culture Technology (GSCT), KAIST. His research interests include modeling and animation of digital humans, avatars, and virtual characters for VR/AR, telepresence, computer game, and computer animation. He received the Ph.D. degree in Computer Science from University of California, Los Angeles, USA, in 2008. He is currently a Professor with the Graduate School of Culture Technology at KAIST. Prior to joining KAIST in 2013, he was an Assistant Professor at Gwangju Institute of Science and Technology (GIST) from 2010 to 2013, and a Postdoctoral researcher at UCLA from 2008 to 2009, and at Honda Research Institute, CA, from 2009 to 2010. He was awarded Outstanding Ph.D. in Computer Science and Northrop Grumman Outstanding Graduate Research Award from UCLA in 2009, Achievement Award from Korea Society of Computer Graphics in 2016 and 2020, Research Innovation Award from KAIST College of Liberal Arts and Convergence Science in 2017 and 2019.

Submission Instructions

XRMemory'25 welcomes a range of paper types (2 to 6 pages, excluding references), including:

Position Papers that share unique insights, opinions, and forward-thinking ideas
Research Papers that present the latest advancements, backed by results, including applications and systems related to spatial memory in XR
Survey Papers that provide a comprehensive overview of the current research landscape, highlighting research gaps and future opportunities
Project Papers that describe the approach and objectives of ongoing or planned projects

We invite a diverse range of topics, encompassing various disciplines and perspectives, to contribute to XRMemory'25, including but not limited to:

AI-driven methods for recording and enhancing spatial experience recall in XR environments.
Cognitive and emotional modeling for interpreting and representing recorded spatial experiences.
Techniques for capturing, reconstructing, and replaying personal experiences in immersive spaces.
Multisensory integration strategies for improving spatial memory retention and replay.
Design principles for virtual environments that support ecologically valid spatial memory research.

Accepted papers will be published in the IEEE Digital Library within the IEEE ISMAR Abstract and Workshop proceedings. At least one author of each accepted paper is required to register, attend, and present their work at XRMemory.

Submission Guidelines

All papers should be original works that are not under review by any other journal or conference. Submissions must be written in English and formatted according to the IEEE Computer Society VGTC (IEEE ISMAR Conference) standards. Templates for LaTeX and Word, as well as a sample PDF, are available at: https://tc.computer.org/vgtc/publications/conference/. Papers should be submitted electronically in the Microsoft CMT service (https://cmt3.research.microsoft.com/XRMemory2025). Each submission will undergo a single-blind review process conducted by the workshop organizers and international program committees.

The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.

Submission Deadlines

All deadlines are 23:59:59 Anywhere on Earth (GMT/UTC-12:00)

Submission deadline: July 19, 2025
Notification of results: August 1, 2025
Camera-ready deadline for inclusion in the IEEE Digital Library: August 15, 2025