- CLIP-Actor-X: Text-driven 4D Human Avatar Generation via Cross-modal Synthesis-through-Optimization (Under review)
- The Devil is in the Details: Simple Remedies for Image-to-LiDAR Representation Learning (ACCV 2024)
- Uni-DVPS: Unified Model for Depth-Aware Video Panoptic Segmentation (RA-L 2024, Oral presentation @ IROS 2024)
- MinDVPS: Minimal Model for Depth-aware Video Panoptic Segmentation (CVPRW 2023)
- CLIP-Actor: Text-Driven Recommendation and Stylization for Animating Human Meshes (ECCV 2022, Qualcomm Innovation Award 2022, AI4CC Workshop @ CVPR 2023)
- Unified 3D Mesh Recovery of Humans and Animals by Learning Animal Exercise (BMVC 2021)
Awards & Honors:
- Best Paper Award at Korean Robotics Society Annual Conference (KRoC), 2024
- Minister’s Award at Electronics Times ICT Paper Awards (1st rank; $12,000), 2023
- Winner of Qualcomm Innovation Fellowship Korea ($4,000), 2022
- Outstanding Poster Presentation, IPIU, 2022
- National Science & Technology Scholarship, 2017-2019
- Academic Excellence Scholarship, 2015-2016
Research Experience
Member of AMILab.
Education
Ph.D. student at POSTECH in Convergence IT Engineering, advised by Tae-Hyun Oh; Bachelor's degree in IT Engineering from Sookmyung Women's University; Exchange student at the University of Toronto.
Background
Research interests: computer vision and machine learning, specializing in understanding the 3D world through 2D visual cues. Recently, her research interests are on perceiving 3D humans in the scene.
Miscellany
Academic Services:
- Conference Reviewer: CVPR, NeurIPS, ICCV, ICLR, ICRA