Seunghun Lee

I am an AI Engineer at Huawei Singapore. I received my Ph.D. and B.S. degrees from DGIST (Daegu Gyeongbuk Institute of Science and Technology), where I was advised by Prof. Sunghoon Im.

My research trajectory is built on a deep understanding of visual and multimodal systems. During the early stages of my doctoral studies, I focused on Domain Adaptation. I subsequently expanded my expertise into Video Perception, tackling tasks such as Video Instance Segmentation (VIS), Referring Video Object Segmentation (RVOS), and Moment Retrieval. Currently, my research interests are centered on Vision-Language-Action (VLA) modeling, exploring the intersection of multimodal reasoning and actionable intelligence.

CV Google Scholar GitHub LinkedIn
Seunghun Lee

Recent Updates

  • Selected as a Top Reviewer at NeurIPS 2026.
  • Joined Huawei Singapore Research Center as an AI Engineer.
  • Started as a postdoctoral researcher at DGIST.
  • Two papers accepted to CVPR 2026.
  • Received my Ph.D. degree from DGIST.

Experience

AI Engineer

Huawei Singapore Research Center, Singapore

Postdoctoral Researcher

DGIST, Daegu, Republic of Korea

Research Intern

LG AI Research

Visiting Researcher

Stanford University, California, US, with Prof. Ehsan Adeli

Education

M.S.-Ph.D. Course, Electrical Engineering and Computer Science

DGIST, Daegu, Republic of Korea

B.S., Convergence Engineering

DGIST, Daegu, Republic of Korea

Preprint

* Co-first author† Corresponding author

RAIN: Robotic Task Generalization with Region-Aware Interaction Networks

Seunghun Lee*, Kyuhwan Shim*, Yeogyeong Song*, Seung Hwan Kim, Sihaeng Lee, Byoung-Tak Zhang, Sunghoon Im†, Gunhee Lee†

Paper Project Code RAIN GitHub stars

Temporal Grounding as A Learning Signal for Referring Video Object Segmentation

Seunghun Lee*, Jiwan Seo*, Jeonghoon Kim*, Sungho Moon*, Siwon Kim, Haeun Yun, Hyogyeong Jeon, Wonhyeok Choi, Jaehoon Jeong, Zane Durante, Sang Hyun Park, Sunghoon Im†

Paper Project Code
EviSplat multi-view evidence overview

EviSplat: Preserving Multi-View Evidence in 3D Gaussian Splatting for Open-Vocabulary Segmentation

Sungho Moon, Kota Shimomura, Junwoo Park, Wonhyeok Choi, Seunghun Lee, Takayoshi Yamashita, Sunghoon Im

Paper

Publications

* Co-first author† Corresponding author

CVA framework overview
CVPR 2026

CVA: Context-aware Video-text Alignment for Video Temporal Grounding

Sungho Moon*, Seunghun Lee*, Jiwan Seo, Sunghoon Im†

Paper Project
TaskForce framework overview
CVPR 2026

TaskForce: Cooperative Multi-agent Reinforcement Learning for Multi-task Optimization

Wonhyeok Choi, Kyumin Hwang, Jihun Park, Kyoungmin Lee, Seunghun Lee, Jaeyeul Kim, Minwoo Choi, Sunghoon Im†

Paper
ICCV 2025

Latest Object Memory Management for Temporally Consistent Video Instance Segmentation

Seunghun Lee, Jiwan Seo, Minwoo Choi, Kiljoon Han, Jaehoon Jeong, Zane Durante, Ehsan Adeli†, Sang Hyun Park, Sunghoon Im†

Paper Project Code LOMM GitHub stars
ICCV 2025

CAVIS: Context-Aware Video Instance Segmentation

Seunghun Lee*, Jiwan Seo*, Kiljoon Han, Minwoo Choi, Sunghoon Im†

Paper Project Code CAVIS GitHub stars
Style-Editor project thumbnail
CVPR 2025Highlight· Top 3.7%

Style-Editor: Text-driven object-centric style editing

Jihun Park*, Jongmin Gim*, Kyoungmin Lee*, Seunghun Lee, Sunghoon Im†

Paper Project Code Style-Editor GitHub stars
Offline-to-Online Knowledge Distillation project thumbnail
WACV 2024Oral· Top 2.6%

Offline-to-Online Knowledge Distillation for Video Instance Segmentation

Hojin Kim, Seunghun Lee, Hyeon Kang, Sunghoon Im†

Paper
Class-Wise Attribute Transfer project thumbnail
BMVC 2023

Multi-Target Domain Adaptation with Class-Wise Attribute Transfer in Semantic Segmentation

Changjae Kim, Seunghun Lee, Sunghoon Im†

Paper
ADAS project thumbnail
CVPR 2022

ADAS: A Direct Adaptation Strategy for Multi-Target Domain Adaptive Semantic Segmentation

Seunghun Lee, Wonhyeok Choi, Changjae Kim, Minwoo Choi, Sunghoon Im†

Paper Code ADAS GitHub stars
DRANet project thumbnail
CVPR 2021

DRANet: Disentangling Representation and Adaptation Networks for Unsupervised Cross-Domain Adaptation

Seunghun Lee, Sunghyun Cho, Sunghoon Im†

Paper Code DRANet GitHub stars

Honor & Award

  • Top Reviewer, NeurIPS 2026
  • 1st Place, 4th GroceryVision Challenge, Spatial-Temporal Action Localization, ICCV 2025
  • DGIST Post-Graduate Research Abroad Award, DGIST
  • Participation Prize, 30th HumanTech Paper Award, Samsung Electronics Co., Ltd.
  • Outstanding Researcher Award, 2022 DGIST Student Conference, EECS
  • Participation Prize, 28th HumanTech Paper Award, Samsung Electronics Co., Ltd.
  • 3rd Place Prize, ICT Paper Contest

Academic Activities

Reviewer

CVPR, ICCV, NeurIPS
ECCV, ICML, ICLR, AAAI
Pacific Graphics

Advanced Studies

International Computer Vision Summer School (ICVSS)