Seunghyeon Lee — accepted AI-conference papers

2026

Affostruction: 3D Affordance Grounding with Generative Reconstruction

Chunghyun Park, Seunghyeon Lee, Minsu Cho

CVPR 2026

This paper addresses the problem of affordance grounding from RGBD images of an object, which aims to localize surface regions corresponding to a text query that describes an action on the object. While existing methods predict affordance regions only on visible surfaces, we propose Affostruction, a

Cited by 0 Source ↗Scholar ↗

2026

EZ-Therm: Effective Zero-Shot Thermal Depth Completion by Adapting Diffusion Priors

Geonhwa Son, Taejoo Kim, Jeongmin Shin, Seunghyeon Lee, Namil Kim, Yukyung Choi

RA-L 2026

For autonomous systems operating in degraded visual environments, robust 3D perception depends on accurate, metric-scale depth prediction. Yet, thermal depth completion—despite being well-suited for such conditions—remains relatively underexplored. Conventional depth completion approaches, which fus

Cited by 0 Source ↗Scholar ↗