Jihun Yoon

About Me

I'm currently working at Hutom as a Lead AI Research Engineer. My research interests include Multimodal Large Language Models (MLLM), world models, and data-efficient learning. Previously, I primarily conducted research on 2D/3D perception from egocentric videos and developed methods for photorealistic synthetic data generation using digital twins. I enjoy collaboration—feel free to reach out!

Research Interests

My research focuses on:

• MLLM
• World Model
• Data Efficient Learning

Projects

Interpretable Surgical Navigation via Vision-Language-Action Model

This project presents an VLA-based surgical navigation system. The system is an interpretable and highly accurate navigation tool capable of real-time decision support and automated view switching, significantly enhancing the safety and efficiency of minimally invasive surgeries.

Publications

• Deep learning-based surgical phase recognition in laparoscopic cholecystectomy
  Authors: Hye Yeon Yang, Seung Soo Hong, Jihun Yoon, et al.
  Journal: Annals of Hepato-Biliary-Pancreatic Surgery (AHBPS 2024)
  [PDF]
• Towards Precise Pose Estimation in Robotic Surgery: Introducing Occlusion-Aware Loss
  Authors: Jihun Park, Jiuk Hong, Jihun Yoon, Bokyung Park, Min-Kook Choi, Heechul Jung
  Conference: Medical Image Computing and Computer-Assisted Intervention (MICCAI 2024)
  [PDF]
• Automated Patient-Specific Pneumoperitoneum Model Reconstruction for Surgical Navigation Systems in Distal Gastrectomy
  Authors: Saebom Shin, Hye-su Jin, Kyungyoon Jung, Bokyung Park, Jihun Yoon, et al.
  Conference: International Workshop on Predictive Intelligence in Medicine (MICCAIW 2024)
  [PDF] [Project Page]
• SP-NAS: Surgical Phase Recognition-Based Navigation Adjustment System for Distal Gastrectomy
  Authors: Hyeongyu Chi, Bogyu Park, Keunyoung Kim, Jiwon Lee, Sungjae Kim, Hyeonu Jeong, Jihun Yoon, et al.
  Conference: International Workshop on Applications of Medical AI (MICCAIW 2024)
  [PDF] [Project Page]
• Exploring video frame redundancies for efficient data sampling and annotation in instance segmentation
  Authors: Jihun Yoon, Min-Kook Choi
  Conference: IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW 2023)
  [PDF] [Code]
• Rethinking Soft Label in Label Distribution Learning Perspective
  Authors: Seungbum Hong, Jihun Yoon, Bogyu Park, Min-Kook Choi
  Preprint: arXiv preprint arXiv (arXiv 2023)
  [PDF]
• Surgical Scene Segmentation Using Semantic Image Synthesis with a Virtual Surgery Environment
  Authors: Jihun Yoon et al.
  Conference: Medical Image Computing and Computer-Assisted Intervention (MICCAI 2022)
  [PDF] [Project Page]
• Self-Supervised Knowledge Transfer via Loosely Supervised Auxiliary Tasks
  Authors: Seungbum Hong, Jihun Yoon, Min-Kook Choi, Junmo Kim
  Conference: IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2022)
  [PDF] [Code]
• hSDB-instrument: instrument localization database for laparoscopic and robotic surgeries
  Authors: Jihun Yoon et al.
  Conference: Medical Image Computing and Computer Assisted Intervention (MICCAI 2021)
  [PDF] [Project Page]
• Semi-supervised object detection with sparsely annotated dataset
  Authors: Jihun Yoon, Seungbum Hong, Min-Kook Choi
  Conference: IEEE International Conference on Image Processing (ICIP 2021)
  [PDF]
• Semi-supervised learning for instrument detection with a class imbalanced dataset
  Authors: Jihun Yoon, Jiwon Lee, SungHyun Park, Woo Jin Hyung, Min-Kook Choi
  Conference: International Workshop on Interpretability of Machine Intelligence in Medical Image Computing (MICCAIW 2020)
  [PDF]
• Real-time optical flow estimation on embedded system
  Authors: Jihun Yoon, Min-Kook Choi, Sangchul Lee
  Conference: 28th Workshop on Image Processing and Image Understanding (IPIU 2016)