Seungjae Kim

I am a M.S. candidate at Visual Science Lab, Kyung Hee University, advised by Prof. MyeongAh Cho.
I build robust visual perception systems that track, detect, and reconstruct dynamic 3D environments.

Video Understanding 2D/3D Perception Novel View Synthesis
email cv github scholar
Seungjae Kim

News

2026.02 🏆 Best Presentation Award at KSC 2025
2025.11 Paper accepted at AAAI 2026 (Acceptance rate: 17.6%)

Publications

PlugTrack
PlugTrack: Multi-Perceptive Motion Analysis for Adaptive Fusion in Multi-Object Tracking
Seungjae Kim, SeungJoon Lee, MyeongAh Cho
AAAI, 2026
arXiv code project page

A plug-and-play motion prediction framework that adaptively fuses Kalman filter and any data-driven motion predictors for robust multi-object tracking.

Based on Jon Barron