강연 초록

In his essay "The Bitter Lesson," Richard Sutton argued that general methods leveraging computation ultimately outperform hand-crafted ones. In this talk, I share my own version of this lesson, learned the hard way over a decade in computer graphics. Where does the bitter lesson apply to graphics, and where does it not? I argue that the answer depends on whether a problem genuinely requires 3D, physics, and control, or if it simply produces 2D pixels. For graduate students starting their research today, my goal is not pessimism but a sharper compass: showing which directions will compound in value over a decade, and which will be quietly subsumed by the next scale-up.

연사 소개

• Meta 연구원 (2020 - 현재)
• KAIST 전산학 박사 (2019)

강연 초록

최근 컴퓨터그래픽스는 콘텐츠 제작을 넘어, 물리 세계와 상호작용하는 ‘피지컬 AI’를 육성하는 핵심 인프라로 진화하고 있습니다. 본 강연에서는 NC AI의 멀티모달 생성형 AI 기술이 로봇의 두뇌를 학습시키는 ‘월드 모델 및 디지털 트윈' 기술로 확장되는 여정을 소개합니다. 특히 로봇이 물리 법칙을 안전하게 학습할 수 있는 가상 환경인 '로봇 운동장'을 구축하기 위해, 당사가 고정밀 스캔 기술, 3D 생성 AI 모델, 뉴럴 렌더링을 어떻게 융합하고 있는지 공유합니다. 이를 통해 가상 세계를 창조하던 그래픽스 기술이 현실의 물리적 지능을 가속화하는 핵심 동력이 되는 미래 비전을 제시하고자 합니다.

연사 소개

• (2015) Ph.D. in Electrical and Electronic Engineering, Korea University
• (2015~17) LG Electronics VIS Lab, Senior Research Engineer
• (2018~22) NCSOFT AI Center Generation Team Leader
• (2022~24) NCSOFT AI Center Vision AI Lab Leader
• (2024~25) NCSOFT AI Tech Center, Head of Division
• (2025~현재) NC AI Chief Technology Officer

강연 초록

The real world is full of diversity, including complex geometry, vivid appearance, and intricate lighting. In applications like autonomous driving and industrial design, building realistic 3D digital representations of the physical world is often essential. This requires us to acquire both geometry and materials from real scenes. The key challenges are how to represent geometry and materials effectively, and how to impose physically meaningful constraints. In this report, I will present our team's work on geometry and material reconstruction over the past two years.

연사 소개

• Professor at Nanjing University
• Research interests include realistic rendering, neural rendering and 3D vision
• B.S. in 2009 and Ph.D. in 2014 from Shandong University
• Joint Ph.D. student at Telecom ParisTech from 2012 to 2014
• Postdoctoral fellow with INRIA (Grenoble) from 2015 to 2017
• Contributed to Disney Infinity at Studio Gobo (UK)
• Published more than thirty papers in ACM TOG, SIGGRAPH (Asia), CVPR, ICCV and other top-tier venues
• Serves on the editorial boards of ACM TOG, Journal of Computer Graphics Techniques and Computer Graphics Forum
• Program Co-Chair of EGSR 2026 and member of the SIGGRAPH 2025 Sorting Committee