Hyun-Kurl Jang

I am a 4th-year Ph.D. student advised by Prof. Kuk-Jin Yoon at the Visual Intelligence Lab (VILab), Korea Advanced Institute of Science and Technology (KAIST).

My research focuses on enabling vision models to continuously learn and adapt beyond the lab. I explore how models can leverage unlabeled real-world data to improve their own representations and predictions over time. My goal is to build robust perception systems across images, videos, and 3D environments that continuously evolve and become more reliable as they encounter the real world.

Publications

* denotes equal contribution.

• 

  Preference-Guided Adaptation for Open-Vocabulary Semantic Segmentation via Prompt Disagreement

  H. Jang, J. Kim, K. Yoon

  NeurIPS 2026

  paper code

• 

  Distill Once, Adapt Life-Long: Exploring Dataset Distillation for Continual Test-Time Adaptation

  H. Jang*, J. Kim*, H. Kweon*, K. Yoon

  ECCV 2026

  paper code

• 

  FrozenDrive: Zero-Shot Text-Guided Driving Scene Generation and Data Augmentation with Parameter-Free Frozen Diffusion Model

  Y. Jeong*, H. Kim*, D. We*, S. Song*, J. Yang*, H. Jang, Y. Yoon, S. Jung, K. Yoon

  ECCV 2026

  paper code

• 

  Test-Time Training for LiDAR Semantic Segmentation under Corruption via Geometric Inlier Discrimination

  H. Kim*, H. Jang*, K. Yoon

  CVPR 2026

  paper code

• 

  TALoS: Enhancing Semantic Scene Completion via Test-time Adaptation on the Line of Sight

  H. Jang*, J. Kim*, H. Kweon*, K. Yoon

  NeurIPS 2024

  paper code

• 

  GMT: Enhancing Generalizable Neural Rendering via Geometry-Driven Multi-Reference Texture Transfer

  Y. Yoon*, H. Jang*, K. Yoon

  ECCV 2024

  paper

• 

  Event-based Video Frame Interpolation with Cross-Modal Asymmetric Bidirectional Motion Fields

  T. Kim, Y. Chae, H. Jang, K. Yoon

  CVPR 2023 Highlight · Top 2.5%

  paper code

Honors & Awards

• 2025Selected as a NeurIPS 2026 Top Reviewer
• 2025Selected as a NeurIPS 2025 Top Reviewer
• 2023Gold Prize, Best Paper Awards, IPIU 2023 (35th Workshop on Image Processing and Image Understanding)

I have served as a reviewer for CVPR, ICCV, ECCV, and NeurIPS.

Projects

• Autonomous Driving in Unstructured Off-Road Environments

  Hanwha Aerospace

  2025–

• Data Augmentation & Sensor Fusion for Robust Autonomous Driving

  42dot

  2025

• Advisor for 2025 HMG Autonomous Driving Challenge

  Hyundai Motor Group

  2024

• Intelligent X-ray Luggage Scanning System

  Korea Customs Service

  2022–24