Events

Date: 5 December 2025   Time: 11:00 - 12:00     

Location: Peter Landin Building 302 (PL 302)

Centre for Intelligent Sensing seminar by Dr Jongyoo Kim and Dr Jungin Park (Yonsei University)
When: 11am-12pm, Friday 5 December
Where: Peter Landin Building 302 (PL 302)

Speaker #1: Dr Jongyoo Kim, Assistant Professor @ Yonsei University, South Korea
(Web: jongyookim.github.io/)
Title: Visual Intelligence: Beyond 2D Generation towards 3D Understanding

Abstract:
This presentation covers a spectrum of advanced vision tasks. We start with the evolution of generative models, discussing inpainting, guidance mechanisms for enhanced quality in diffusion and flow models, and watermarking for model security. Expanding into the third dimension, we introduce Text-to-3D synthesis and delve into the complexities of 3D human-centric perception, including pose/hand tracking and human-object reconstruction in wild scenarios. Join us to see how these diverse technologies converge to shape the future of visual AI.

Bio:
Jongyoo Kim is currently an Assistant Professor in the School of Electrical and Electronic Engineering at Yonsei University. Before joining Yonsei, he was a Senior Researcher at Microsoft Research Asia from 2018 to 2024. He also serves as an Associate Editor for the IEEE Transactions on Multimedia.

Speaker #2: Dr Jungin Park, Research Professor @ Yonsei University, South Korea
(Web: park-jungin.github.io/)
Title: Bootstrapping your own views: Masked ego-exo modeling for fine-grained view-invariant video representations

Abstract:
This presentation introduces a novel fine-grained view-invariant video representation learning from unpaired ego-exo videos, called Bootstrap Your Own Views (BYOV). We highlight the importance of capturing the compositional nature of human actions as a basis for robust cross-view understanding. To this end, we introduce a masked ego-exo modeling that promotes both causal temporal dynamics and cross-view alignment. Specifically, self-causal masking and cross-view masking predictions are learned concurrently to facilitate view-invariant and powerful representations across viewpoints.

Bio:
Jungin Park is currently a Research Professor in the School of Electrical and Electronic Engineering at Yonsei University. He received the B.E. and Ph.D. degrees in Electrical and Electronic Engineering from Yonsei University, Seoul, Korea, in 2018 and 2024, respectively. His current research interests include computer vision, video understanding, and multimodal learning.