MIRA

2nd Workshop on

Multispectral Imaging for Robotics and Automation

ICCV 2025 Workshop

ICCV 2025


Introduction

Multispectral imaging is revolutionizing the fields of robotics and automation by providing richer information beyond the visible spectrum. Traditional RGB cameras capture only a narrow band of the electromagnetic spectrum, limiting the data available for computer vision systems. Multispectral cameras expand this capability by sensing light across a broader range of wavelengths, including infrared, ultraviolet, and other portions of the spectrum invisible to the human eye.

This additional spectral information unlocks powerful new applications in robotics and automation. Multispectral data can be used for enhanced material classification, detecting various objects, identifying chemical signatures, and perceiving environmental factors like moisture and temperature. Additionally, in autonomous driving, multispectral imaging allows vehicles to detect lane markings better, read traffic signals, and identify obstacles in challenging conditions like adverse weather situations and darkness. Such capabilities have transformative potential for industrial inspection, agricultural automation, search and rescue operations, self-driving cars, and countless other domains.

The Multispectral Imaging for Robotics and Automation (MIRA) workshop aims to bring together leading researchers exploring this emerging interdisciplinary area at the intersection of multispectral imaging, computer vision, robotics, and automation.

Join us to discuss the latest breakthroughs, share cutting-edge research, and forge new collaborations driving innovation in this exciting field!


Call For Papers

We invite original, unpublished work that advances spectral sensing, processing, and applications in robotic and automated systems. Submissions may include full papers (up to 8 pages). Relevant topics include but are not limited to:

  • Sensor & Fusion: Multispectral data acquisition, calibration, and fusion with LiDAR/RADAR/thermal/event cameras
  • Perception & Vision: Detection, segmentation, SLAM, and through-obstacle imaging in challenging conditions (weather, low-light, underwater)
  • Learning & Adaptation: Self-/unsupervised learning, domain adaptation, generative methods, explainability, and few/zero-shot learning for spectral data
  • Applications: Industrial inspection, agriculture, autonomous vehicles (ground/AUV/UAV), medical robotics, and human–robot systems
  • Systems & Data: Edge computing, real-time processing, dataset curation, benchmarks, and standardization
  • Emerging Directions: Vision-language integration, digital twins, swarm spectral perception, and ethical/environmental considerations
  • ...


Paper Submission Guidelines:

  • All submissions will be collected electronically via workshop's OpenReview portal: https://openreview.net/group?id=thecvf.com/ICCV/2025/Workshop/MIRA.
  • Please note, all submitting authors must have an Openreview profile, if you do not have one, please create one as soon as possible by using an institutional email. Papers that have been previously published or are currently under review elsewhere will not be accepted.
  • Submissions should adhere to the formatting standards outlined by the International Conference on Computer Vision (ICCV), which can be found at https://iccv.thecvf.com/Conferences/2025/AuthorGuidelines.
  • The paper should not exceed 8 pages, including figures and tables. However, additional pages are permissible solely for references. Submitted papers must be anonymized and use the ICCV LaTeX template ("ICCV 2025 Author Kit"). Papers that are not properly anonymized, or do not use the template, or have more than eight pages (excluding references) will be rejected without review.
  • All accepted papers are going to be presented in poster session and will be published in the ICCV 2025 Workshop proceedings.
  • Selected papers will be invited to submit an extended version to "Embodied Multi-Modal Data Fusion for Robot Continuous Perception" Special Issue of the Pattern Recognition Letters.


Important Dates

Call for papers announced May 10, 2025
Paper submission deadline July 1, 2025 (11:59 PM Honolulu time)
Notifications to accepted papers July 10, 2025
Camera-ready submission deadline August 15, 2025 (11:59 PM Honolulu time)
Workshop date October 19, 2025


Keynote Speakers




Ko Nishino

Kyoto University



Björn Ommer

Ludwig Maximilian University of Munich (LMU)



Xiaodan Liang

Sun Yat-Sen University



Lin Wang

Nanyang Technological University (NTU)



Tianfan Xue

Chinese University of Hong Kong



Ayoung Kim

Seoul National University (SNU)



Siheng Chen

Shanghai Jiao Tong University


Organizers

Shiho Kim
Yonsei University
Yağız Nalçakan
Yonsei University
Rui (Ranger) Fan
Tongji University
Kailun Yang
Hunan University
Yalın Baştanlar
Izmir Institute of Technology
Jun Won Choi
Seoul National University
Ömer Şahin Taş
Karlsruhe Institute of Technology (KIT) & Forschungszentrum Informatik (FZI)
Ukcheol Shin
Carnegie Mellon University
Michal Kovac
Slovak University of Technology in Bratislava

Program & Technical Committee

  • Wei Ye - Tongji University
  • Yi Feng - Tongji University
  • Mengtan Zhang - Tongji University
  • Hongbo Zhao - Tongji University
  • Youngwan Jin - Yonsei University
  • Incheol Park - Yonsei University
  • Hyeongjin Ju - Yonsei University
  • Sanghyeop Yeo - Yonsei University


Contact

To contact the organizers please use multispectral4ra@outlook.com and ynalcakan@yonsei.ac.kr


Acknowledgments

Thanks to visualdialog.org and L3DS workshop team for the webpage format.