Jinhyeok Choi

prof_img_.png

I am a second-year Ph.D. student at KAIST AI, Seoul, South Korea, advised by Prof. Seungryong Kim. My research lies at the intersection of 3D/4D computer vision and generative models, with a particular interest in leveraging 3D/4D representations for geometrically consistent multi-view and video generation, as well as exploring generative approaches to 3D/4D reconstruction and creation, from dynamic objects to large-scale environments. Currently, I focus on advancing video world models to maintain long-term scene consistency and capture the underlying 3D geometry of the world, while enabling efficient generation and memory usage.

Prior to joining KAIST, I received the B.S. degree in Computer Science and Engineering at Korea University, Seoul, South Korea.

News

Nov. 2026 🚀 My two recent papers, L3G-World and World Observer are now out! Check out the project pages for details.
Jun. 2026 🚀 My two recent papers, MORPHOS and GeoFace are now out! Check out the project pages for details.
Jun. 2026 🎉 Two papers accepted to ECCV 2026! Check Seoul World Model and Geometric Latent Diffusion.
Feb. 2026 🎉 Two papers accepted to CVPR 2026, including my first lead-author paper CAMEO!
Sep. 2025 Started my Ph.D studies at KAIST AI !

Selected publications

  1. arXiv 2026
    l3g-world.png
    L3G-World: Real-Time Infinite World Generation with Latent 3D Gaussian Memory
    Jinhyeok Choi*, JoungBin Lee*, Jongmin Lee, Minkyeong Jeon, Jiyoung Kim, Kaeun Han, Yoonjin Yoon, and Seungryong Kim
    arXiv preprint, 2026
    arXiv PDF Project Page Code
  2. arXiv 2026
    world_observer.gif
    World Observer: Joint Actor-Observer Generation for Persistent World Modeling
    Hyunwook Choi, Dahyun Chung, Hyunsung Kim, Siyoon Jin, Jinhyeok Choi, Junyoung Seo, and Seungryong Kim
    arXiv preprint, 2026
    arXiv PDF Project Page Code
  3. arXiv 2026
    morphos.png
    MORPHOS: Autoregressive 4D Generation with Temporal Structured Latents
    Minkyung Kwon*, Jinhyeok Choi*, Youngjin Shin, Jaeyeong Kim, JongMin Lee, and Seungryong Kim
    arXiv preprint, 2026
    arXiv PDF Project Page Code
  4. arXiv 2026
    geoface.png
    GeoFace: Consistent Multi-View Face Generation with Geometry-Constrained Diffusion
    Yeji Choi, Jinhyeok Choi, Jaewon Min, Minkyung Kwon, Jin Hyeon Kim, and Seungryong Kim
    arXiv preprint, 2026
    arXiv PDF Project Page Code
  5. ECCV 2026
    swm.gif
    Grounding World Simulation Models in a Real-World Metropolis
    Junyoung Seo*, Hyunwook Choi*, Minkyung Kwon, Jinhyeok Choi, Siyoon Jin, Gayoung Lee, Junho Kim, JoungBin Lee, Geonmo Gu, Dongyoon Han, Sangdoo Yun, Seungryong Kim, and Jin-Hwa Kim
    ECCV, 2026 Spotlight
    arXiv PDF Project Page Code
  6. ECCV 2026
    gld.png
    Repurposing Geometric Foundation Models for Multi-view Diffusion
    Wooseok Jang, Seonghu Jeon, Jisang Han, Jinhyeok Choi, Minkyung Kwon, Seungryong Kim, Saining Xie, and Sainan Liu
    ECCV, 2026 Oral
    arXiv PDF Project Page Code
  7. CVPR 2026
    cameo.png
    CAMEO: Correspondence-Attention Alignment for Multi-View Diffusion Models
    Minkyung Kwon*, Jinhyeok Choi*, Jiho Park*, Seonghu Jeon, Jinhyuk Jang, Junyoung Seo, Minseop Kwak, Jin-Hwa Kim, and Seungryong Kim
    CVPR, 2026
    arXiv PDF Project Page Code
  8. CVPR 2026
    apple.png
    Attribute-Preserving Pseudo-Labeling for Diffusion-Based Face Swapping
    Jiwon Kang, Yeji Choi, JoungBin Lee, Wooseok Jang, Jinhyeok Choi, Taekeun Kang, Yongjae Park, Myungin Kim, and Seungryong Kim
    CVPR, 2026
    arXiv PDF Project Page Code

Education

KAIST Ph.D. in Artificial Intelligence
KAIST AI, Seoul, South Korea
Advised by Prof. Seungryong Kim 		Sep. 2025 - Present
Korea Univ B.S. in Computer Science and Engineering
Korea University, Seoul, South Korea
Graduated with Great Honor (GPA: 4.42/4.5) 		Mar. 2022 - Aug. 2025