Daehee Park, Ph.D.
Assistant Professor
Depart. of Electrical Engineering and Computer Science (EECS)
Depart. Interdisciplinary Studies of Artificial Intelligence (AI)
Daegu Gyeongbuk Institute of Science & Technology (DGIST)
Lab: Intelligent Systems and Learning (ISL) Lab
Email: dhpark@dgist.ac.kr
News
(Jul 2025) I gave an invited talk at the 2025 Joint Summer School of the Visual Understanding and Image Processing Societies [Link]
(Jun 2025) Two papers on motion planning and diffusion generative method were accepted at ICCV 2025 (one elected as a hightlight✨)
(Jun 2025) One paper on diffusion motion planning was accepted at IROS 2025
(Jun 2025) I gave an invited talk at the 2025 Workshop on Embodied AI / Physical AI [News]
(May 2025) I gave an invited talk at the Daegu AI Researchers Forum [Link]
(May 2025) I gave an invited talk on the top-tier AI conference session at the Spring Conference of the Korean Society of Automotive Engineers [Link]
(Feb 2025) One paper on multi-modal motion forecasting was accepted at CVPR 2025
(Feb 2025) I joined DGIST as an Assitant Professor at depart. of EECS
(Jan 2025) I am selected as an winner of Qualcomm Innovation Fellowship Korea (QIFK) 2024
(Oct 2024) I am selected as a finalist of Qualcomm Innovation Fellowship Korea (QIFK) 2024
(Jul 2024) One paper on semantic segmentation was accepted at ECCV 2024
(Mar 2024) I joined Qualcomm Research as an Ph.D. intern
(Feb 2024) Two papers on test-time adaptation and human motion forecasting were accepted to CVPR 2024 (one elected as a hightlight✨)
(Dec 2023) One paper on domain adaptation was accepted at AAAI 2024
(Jan 2023) One paper on trajectory prediction was accepted at ICLR 2023
(Oct 2022) I obtained sponsorship program from Samsung Advanced Institute of Technology (SAIT)!
(Jul 2022) One paper on image generation was accepted at ECCV 2022
(Feb 2022) I won 1st place in the best paper awards during IPIU 2022!
(Jun 2021) One paper on semantic segmentation was accepted at ICCV 2021
(Jun 2021) The paper at RA-L was presented at ICRA 2021
(May 2021) I joined Naver Labs as an Ph.D. intern
(Dec 2020) One paper on object detection was accepted at RA-L
Experience
Feb. 2025 ~ Present, Assistant Professor, DGIST EECS
Apr. 2024 ~ Sep. 2024, Research Intern, QCT Auto AD, Qualcomm
Jun. 2021 ~ Aug. 2021, Research Intern, Autonomous Driving Team, Naver Labs
Education
2025, KAIST, Ph.D. in Mechanical Engineering (Supervisor: Prof. Kuk-Jin Yoon)
2020, KAIST, M.S. in Mechanical Engineering (Supervisor: Yong-Hwa Park)
2018, KAIST, B.S. in Mechanical Engineering
Publications
(c: conference, j: journal, p: preprint, *: equal contribution)
[c13] Generative Active Learning for Long-tail Trajectory Prediction via Controllable Diffusion Model
Daehee Park, Monu Surana, Pranav Desai, Ashish Mehta, Reuben MV John, Kuk-Jin Yoon
International Conference on Computer Vision (ICCV), 2025
(Collaboration with Qualcomm, US)
[c12] Interaction-Merged Motion Planning: Effectively Leveraging Diverse Motion Datasets for Robust Planning
Giwon Lee*, Wooseong Jeong*, Daehee Park, Jaewoo Jeong, Kuk-Jin Yoon
International Conference on Computer Vision (ICCV), Hightlight ✨, 2025
[c11] Non-differentiable Reward Optimization for Diffusion-based Autonomous Motion Planning
Giwon Lee*, Daehee Park*, Jaewoo Jeong*, Kuk-Jin Yoon
International Conference on Intelligent Robots and Systems (IROS), 2025
[p1] Self-Supervised 3D Occupancy Prediction with Temporal Consistency
Jae-Seok Jeong, Sung-Hoon Yoon, Daehee Park, Kuk-Jin Yoon
preprint, 2024
[c5] Leveraging Future Relationship Reasoning for Vehicle Trajectory Prediction
Daehee Park, Hobin Ryu, Yunseo Yang, Jegyeong Cho, Jiwon Kim, Kuk-Jin Yoon
International Conference on Learning Representations (ICLR), 2023
(1st rank in nuScenes prediction challenge)
(Collaboration with Naver Labs)
[paper]
[c3] Unlocking the Potential of Ordinary Classifier: Class-specific Adversarial Erasing Framework for Weakly Supervised Semantic Segmentation
Hyeokjun Kweon*, Sung-Hoon Yoon*, Hyeonseong Kim, Daehee Park, Kuk-Jin Yoon
Proceedings of the IEEE/CVF international conference on computer vision (ICCV), 2021
[paper]
[j1,c2] Identifying Reflected Images from Object Detector in Indoor Environment Utilizing Depth Information
Daehee Park, Yong-Hwa Park
IEEE Robotics and Automation Letters (RA-L), 2020 & IEEE International Conference on Robotics and Automation (ICRA), 2021
[paper]
[c1] A Scanning 3D Sensor and Its Object Recognition for Autonomous Robots
Joon-Oh Shin, In-Gyu Jang, Dae-Hee Park, Yong-Hwa Park
SPIE MOEMS and Miniaturized Systems XVIII, 2019
[paper]
