Jeongho Min | Lumir Inc.

About Me

I am an AI Researcher at Lumir Inc. (Military Service), working on computer vision and remote sensing applications. I received my M.S. in Artificial Intelligence from Ulsan National Institute of Science and Technology (UNIST) in 2023, where I worked on improving reference-based super-resolution in remote sensing via domain matching under the supervision of Prof. Jaejun Yoo, and my B.S. in Computer Science and Engineering from Kyungpook National University in 2021.

My research focuses on inverse problems in computer vision, particularly super-resolution (Blind SR, Reference-based SR, SISR), denoising, and generative AI. I also have extensive experience in building LLM-based systems, including RAG pipelines, tool-calling frameworks, and AI agents for geospatial applications.

Research Interests

Computer Vision: Super-resolution (RefSR, Blind SR, SISR), Denoising, Segmentation, Object Detection, Generative AI (GAN, Diffusion)
Remote Sensing: Multi-modal fusion (EO & SAR), Land cover classification, Cross-view retrieval, Domain adaptation
LLM & Agentic Systems: RAG pipelines, Tool calling, Function orchestration, LangChain/LangGraph/Langflow, Semantic Kernel, CrewAI
Deep Learning: Inverse problems, Domain matching, Multi-modal learning

News

[Nov. 2025] 📄 Our paper “From Street to Orbit: Training-Free Cross-View Retrieval via Location Semantics and Language Guidance” is accepted to WACV 2026
[June 2025] 🏆 Finalist / 3rd Place (Test Phase, out of 104 participants) — Multi-modal Aerial View Image Challenge (PBVS @ CVPR 2025), classification track
[Dec. 2024] 🏆 Finalist / 3rd Place (Test Phase, out of 204 participants) — MMSeg-YREB Multimodal Semantic Segmentation Challenge, IEEE WHISPERS 2024
[2024] 🏢 Joined Lumir Inc. as AI Researcher (Military Service), working on GIS-oriented LLM Chatbot and land cover segmentation
[2023] 📄 Our paper “Bridging the Domain Gap: A Simple Domain Matching Method for Reference-Based Image Super-Resolution in Remote Sensing” is published in IEEE GRSL (Impact Factor: 5.343)

Publications

WACV

From Street to Orbit: Training-Free Cross-View Retrieval via Location Semantics and LLM Guidance

Jeongho Min*, Dongyoung Kim, Jaehyup Lee (*First author)

IEEE/CVF Winter Conference on Computer Vision (WACV), 2026.

Project Page

WHISPERS

Gated-MCBAM: Cross-Modal Block Attention Module with Gating Mechanism for Remote Sensing Segmentation

Jeongho Min*, Sungbin Son, Hyeonjin Kim, Seunghyeok Back, Jaejun Yoo, Youngkyun Kong (*First author)

IEEE WHISPERS (Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing), 2024.

Oral Presentation

JSTARS

Multimodal Semantic Segmentation in Yangtze River Economic Belt: Outcome of the 2024 IEEE WHISPERS MMSeg-YREB Challenge

Jiepan Li, He Huang, Yu Xia, Xian Xu, Yingxin Wu, Qi Guo, Yinhe Liu, Yanfei Zhong, Jeongho Min, Sungbin Son, Hyeonjin Kim, Jaejun Yoo, Gemine Vivone, Chenyu Li, Wei He, Renwei Dian, Hao Liu, Haipeng Wang, Kan Wei, A. K. Qin, Naoto Yokoya, Shutao Li, Jocelyn Chanussot, Danfeng Hong

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2025.

Impact Factor: 5.3

GRSL

Bridging the Domain Gap: A Simple Domain Matching Method for Reference-Based Image Super-Resolution in Remote Sensing

Jeongho Min*, Yejun Lee, Dongyoung Kim, Jaejun Yoo (*First author)

IEEE Geoscience and Remote Sensing Letters, vol. 21, 2023.

Impact Factor: 4.8, JCR: 9.6%

Awards & Honors

Competition Awards

Finalist / 3rd Place (Test Phase, out of 104 participants) — Multi-modal Aerial View Image Challenge (PBVS @ CVPR 2025), classification track, SAR/EO multi-sensor data, June 2025.

Team:
Jeongho Min*,
Hyeonjin Kim (UNIST),
Jaehyup Lee (KNU),
Prof. Jaejun Yoo (UNIST)
Finalist / 3rd Place (Test Phase, out of 204 participants) — MMSeg-YREB Multimodal Semantic Segmentation Challenge, IEEE WHISPERS 2024.

Team:
Jeongho Min* (Team Leader),
Sungbin Son,
Hyunjin Kim (UNIST),
Seunghyeok Back (GIST)

Patents

Jaejun Yoo, Jeongho Min, Yejun Lee, Dongyoung Kim, "Method for Enhancing Image Resolution and Computing Device Performing the Same", KR Patent No. 10-2770741, Registered on Feb 17, 2025. (Licensed by UNIST and officially transferred to D**p** Co., Ltd. through a formal technology transfer agreement.)