avatar

Wonseok Roh

Ph.D. Student
Korea University
paulroh (at) korea.ac.kr


About Me

I am a Ph.D. student in the Department of Artificial Intelligence at Korea University, studying under the guidance of Prof. Sangpil Kim in the Computer Vision Lab. My research primarily focuses on deep learning-based Computer Vision, with a particular emphasis on advancing the interpretation of real-world scenarios across both 2D and 3D domains. Specifically, I aim to develop innovative methods that bridge theoretical advancements and practical applications in visual understanding.
As part of an ongoing research project, I am collaborating with Innfan Yoo and Andreas Lugmayr from Google, focusing on generalizable 3D scene reconstruction in monocular settings.

Research Interests

News 🎉

Publications

  1. ICCV 2025
    CATSplat: Context-Aware Transformer with Spatial Guidance for Generalizable 3D Gaussian Splatting from A Single-View Image
    Wonseok Roh, Hwanhee Jung, Jong Wook Kim, Seunggwan Lee, Innfarn Yoo, Andreas Lugmayr, Seunggeun Chi, Karthik Ramani, Sangpil Kim
    International Conference on Computer Vision (ICCV), 2025.

  2. CVPR 2025
    Insightful Instance Features for 3D Instance Segmentation
    Wonseok Roh, Hwanhee Jung, Giljoo Nam, Dong In Lee, Hyeongcheol Park, Sang Ho Yoon, Jungseock Joo, Sangpil Kim
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025.

  3. CVPR 2024
    Edge-Aware 3D Instance Segmentation Network with Intelligent Semantic Prior
    Wonseok Roh, Hwanhee Jung, Giljoo Nam, Jinseop Yeom, Hyunje Park, Sang Ho Yoon, Sangpil Kim
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024.

  4. AAAI 2024
    CMDA: Cross-Modal and Domain Adversarial Adaptation
    for LiDAR- Based 3D Object Detection
    Gyusam Chang*, Wonseok Roh*, Sujin Jang, Dongwook Lee, Daehyun Ji, Gyeongrok Oh, Jinsun Park, Jinkyu Kim†, Sangpil Kim† (*=Equal contribution)
    The 38th Annual AAAI Conference on Artificial Intelligence (AAAI), 2024.

  5. BMVC 2023
    Functional Hand Type Prior for 3D Hand Pose Estimation and Action Recognition from Egocentric View Monocular Videos
    Wonseok Roh, Seung Hyun Lee, Won Jeong Ryoo, Jakyung Lee, Gyeongrok Oh, Sooyeon Hwang, Hyung-gun Chi, Sangpil Kim
    British Machine Vision Conference (BMVC), 2023.
    Oral Presentation

  6. Information Sciences
    Self-supervised Multimodal Graph Convolutional Network
    for Collaborative Filtering
    Sungjune Kim, Seongjun Yun, Jongwuk Lee, Gyusam Chang, Wonseok Roh, Daneung Sohn, Jungtae Lee, Hogun Park†, Sangpil Kim†
    Information Sciences, 2023, (Impact Factor 8.1).

  7. BMVC 2022
    ORA3D: Overlap Region Aware Multi-view 3D Object Detection
    Wonseok Roh, Gyusam Chang, Seokha Moon, Giljoo Nam, Chanyoung Kim, Younghyun Kim, Sangpil Kim†, Jinkyu Kim†
    British Machine Vision Conference (BMVC), 2022.

  8. CVPR 2022
    Sound-guided Semantic Image Manipulation
    Seung Hyun Lee, Wonseok Roh, Wonmin Byeon, Sang Ho Yoon, Chanyoung Kim, Jinkyu Kim†, Sangpil Kim†
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022.

Patents

  1. U.S. Patent
    Method and apparatus with object detector training
    Sujin Jang, Sangpil Kim, Jinkyu Kim, Wonseok Roh, Gyusam Chang, Dongwook Lee, Daehyun Ji
    U.S. Patent Application No. 18/451,287.

  2. U.S. Patent
    Apparatus and method for detecting an object and a computer readable recording medium therefor
    Younghyun Kim, Seowon Lee, Jinkyu Kim, Sangpil Kim, Wonseok Roh
    U.S. Patent Application No. 18/217,904.

Services

Conference Reviewers


Powered by Jekyll and Minimal Light theme.