Suhwan Cho AI/ML Research Scientist GenGenAI, Seoul, Korea suhwanx [at] gmail.com CV | GitHub | LinkedIn | Google Scholar

Bio

• Technical Lead at GenGenAI, 2025 - Present
• Research Scientist Intern at Adobe Research, 2023
• Ph.D. in EE, Yonsei University, 2020 - 2025
• B.S. in EE, Yonsei University, 2016 - 2020

Research Interests  

• Temporal & Motion-Aware Vision:
  Understanding objects and pixels in video with robust temporal and motion-aware consistency.

• Generative & 3D Vision:
  Leveraging generative priors and 3D representations to enhance video synthesis and scene understanding.

• Multi-Modal & Interactive Vision:
  Bridging vision and language to enable intuitive human-interactive visual control.

• Real-World Vision Systems:
  Translating fundamental computer vision research into robust, scalable, and real-world applicable solutions.

Collaborators

I have been fortunate to work closely with many brilliant researchers, including:

• Yonwoo Choi (GenGenAI)
• Seoung Wug Oh, Joon-Young Lee (Adobe Research)
• Minhyeok Lee, Jungho Lee (Yonsei University)
• And many outstanding colleagues.

I’m always open to research collaborations and academic partnerships. Feel free to reach out!

---

Publications 

^{(* indicates equal contribution)}

2026

• Bidirectional Token-Masking AutoEncoder for Referring Image Segmentation
    Minhyeok Lee, Dogyoon Lee, Jungho Lee, Suhwan Cho, Sangyoun Lee
    Pattern Recognition (PR), 2026

• Generalizing CLIP Prompts for Zero-shot Anomaly Detection
    Donghyeong Kim, Chaewon Park, Suhwan Cho, Hyeonjeong Lim, Minseok Kang, Jungho Lee, Sangyoun Lee
    Pattern Recognition (PR), 2026

• Seen-to-Scene: Keep the Seen, Generate the Unseen for Video Outpainting
    Inseok Jeon, Minhyeok Lee, Seunghoon Lee, Minseok Kang, Suhwan Cho, Sangyoun Lee
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR Findings), 2026

2025

• Find First, Track Next: Decoupling Identification and Propagation in Referring Video Object Segmentation [Code]
    Suhwan Cho*, Seunghoon Lee*, Minhyeok Lee, Jungho Lee, Sangyoun Lee
    The 7th Large-Scale Video Object Segmentation (LSVOS) Workshop @ ICCV, 2025

• DepthFlow: Exploiting Depth-Flow Structural Correlations for Unsupervised Video Object Segmentation [Code]
    Suhwan Cho, Minhyeok Lee, Jungho Lee, Donghyeong Kim, Sangyoun Lee
    The 7th Large-Scale Video Object Segmentation (LSVOS) Workshop @ ICCV, 2025

• TransFlow: Motion Knowledge Transfer from Video Diffusion Models to Video Salient Object Detection [Code]
    Suhwan Cho, Minhyeok Lee, Jungho Lee, Sunghun Yang, Sangyoun Lee
    The 7th Large-Scale Video Object Segmentation (LSVOS) Workshop @ ICCV, 2025

• CoMoGaussian: Continuous Motion-Aware Gaussian Splatting from Motion-Blurred Images [Page]
    Jungho Lee, Donghyeong Kim, Dogyoon Lee, Suhwan Cho, Minhyeok Lee, Wonjoon Lee, Taeoh Kim, Dongyoon Wee, Sangyoun Lee
    IEEE/CVF International Conference on Computer Vision (ICCV), 2025

• CMTM: Cross-Modal Token Modulation for Unsupervised Video Object Segmentation
    Inseok Jeon, Suhwan Cho, Minhyeok Lee, Seunghoon Lee, Minseok Kang, Jungho Lee, Chaewon Park, Donghyeong Kim, Sangyoun Lee
    IEEE International Conference on Image Processing (ICIP), 2025 (Oral)

• Treating Motion as Option with Output Selection for Unsupervised Video Object Segmentation [Code]
    Suhwan Cho, Minhyeok Lee, Jungho Lee, MyeongAh Cho, Seungwook Park, Jaeyeob Kim, Hyunsung Jang, Sangyoun Lee
    IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2025

• Effective SAM Combination for Open-Vocabulary Semantic Segmentation
    Minhyeok Lee, Suhwan Cho, Jungho Lee, Sunghun Yang, Heeseung Choi, Ig-Jae Kim, Sangyoun Lee
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025 (Oral)

• CoCoGaussian: Leveraging Circle of Confusion for Gaussian Splatting from Defocused Images [Page]
    Jungho Lee, Suhwan Cho, Taeoh Kim, Ho-Deok Jang, Minhyeok Lee, Geonho Cha, Dongyoon Wee, Dogyoon Lee, Sangyoun Lee
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025

• Elevating Flow-Guided Video Inpainting with Reference Generation [Code]
    Suhwan Cho, Seoung Wug Oh, Sangyoun Lee, Joon-Young Lee
    AAAI Conference on Artificial Intelligence (AAAI), 2025

• Video Diffusion Models are Strong Video Inpainter [Code]
    Minhyeok Lee, Suhwan Cho, Chajin Shin, Jungho Lee, Sunghun Yang, Sangyoun Lee
    AAAI Conference on Artificial Intelligence (AAAI), 2025

2024

• STATIC : Surface Temporal Affine for TIme Consistency in Video Monocular Depth Estimation
    Sunghun Yang, Minhyeok Lee, Jungho Lee, Suhwan Cho, Sangyoun Lee
    arXiv, 2024

• LSHNet: Leveraging Structure-Prior with Hierarchical Feature Updates for Salient Object Detection in Optical Remote Sensing Images [Code]
    Seunghoon Lee, Suhwan Cho, Chaewon Park, Seungwook Park, Jaeyeob Kim, Sangyoun Lee
    IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2024

• Dual Prototype Attention for Unsupervised Video Object Segmentation [Code]
    Suhwan Cho*, Minhyeok Lee*, Seunghoon Lee, Dogyoon Lee, Sangyoun Lee
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024  

• Guided Slot Attention for Unsupervised Video Object Segmentation [Code]
    Minhyeok Lee, Suhwan Cho, Dogyoon Lee, Chaewon Park, Jungho Lee, Sangyoun Lee
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024  

2023

• Leveraging Spatio-Temporal Dependency for Skeleton-Based Action Recognition [Code]
    Jungho Lee, Minhyeok Lee, Suhwan Cho, Sungmin Woo, Sangyoun Lee
    IEEE/CVF International Conference on Computer Vision (ICCV), 2023

• Adaptive Graph Convolution Module for Salient Object Detection
    Yongwoo Lee, Minhyeok Lee, Suhwan Cho, Sangyoun Lee
    IEEE International Conference on Image Processing (ICIP), 2023

• TSANet: Temporal and Scale Alignment for Unsupervised Video Object Segmentation [Code]
    Seunghoon Lee, Suhwan Cho, Dogyoon Lee, Minhyeok Lee, Sangyoun Lee
    IEEE International Conference on Image Processing (ICIP), 2023

• Two-Stream Decoder Feature Normality Estimating Network for Industrial Anomaly Detection
    Chaewon Park, Minhyeok Lee, Suhwan Cho, Donghyeong Kim, Sangyoun Lee
    IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023

• FAPM: Fast Adaptive Patch Memory for Real-Time Industrial Anomaly Detection [Code]
    Donghyeong Kim, Chaewon Park, Suhwan Cho, Sangyoun Lee
    IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023

• One-Shot Video Inpainting
    Sangjin Lee*, Suhwan Cho*, Sangyoun Lee
    arXiv, 2023

• Treating Motion as Option to Reduce Motion Dependency in Unsupervised Video Object Segmentation [Code]
    Suhwan Cho, Minhyeok Lee, Seunghoon Lee, Chaewon Park, Donghyeong Kim, Sangyoun Lee
    IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023

• Unsupervised Video Object Segmentation via Prototype Memory Network [Code]
    Minhyeok Lee, Suhwan Cho, Seunghoon Lee, Chaewon Park, Sangyoun Lee
    IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023

2022

• Tackling Background Distraction in Video Object Segmentation [Code]
    Suhwan Cho, Heansung Lee, Minhyeok Lee, Chaewon Park, Sungjun Jang, Minjung Kim, Sangyoun Lee
    European Conference on Computer Vision (ECCV), 2022

• SPSN: Superpixel Prototype Sampling Network for RGB-D Salient Object Detection [Code]
    Minhyeok Lee*, Chaewon Park*, Suhwan Cho, Sangyoun Lee
    European Conference on Computer Vision (ECCV), 2022

• Superpixel Group-Correlation Network for Co-Saliency Detection
    Minhyeok Lee, Chaewon Park, Suhwan Cho, Sangyoun Lee
    IEEE International Conference on Image Processing (ICIP), 2022

• Detection-Identification Balancing Margin Loss for One-Stage Multi-Object Tracking
    Heansung Lee, Suhwan Cho, Sungjun Jang, Jungho Lee, Sungmin Woo, Sangyoun Lee
    IEEE International Conference on Image Processing (ICIP), 2022

• Pixel-Level Equalized Matching for Video Object Segmentation
    Suhwan Cho, Woo Jin Kim, MyeongAh Cho, Seunghoon Lee, Minhyeok Lee, Chaewon Park, Sangyoun Lee
    arXiv, 2022

• Unsupervised Video Anomaly Detection via Normalizing Flows with Implicit Latent Features
    MyeongAh Cho, Taeoh Kim, Woo Jin Kim, Suhwan Cho, Sangyoun Lee
    Pattern Recognition (PR), 2022

• Occluded Person Re-Identification via Relational Adaptive Feature Correction Learning
    Minjung Kim, MyeongAh Cho, Heansung Lee, Suhwan Cho, Sangyoun Lee
    IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2022

• Pixel-Level Bijective Matching for Video Object Segmentation [Code]
    Suhwan Cho, Heansung Lee, Minjung Kim, Sungjun Jang, Sangyoun Lee
    IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2022

2020

• CRVOS: Clue Refining Network for Video Object Segmentation
    Suhwan Cho, MyeongAh Cho, Tae-young Chung, Heansung Lee, Sangyoun Lee
    IEEE International Conference on Image Processing (ICIP), 2020

---

Hosted on GitHub Pages — Theme by orderedlist