Soyul Lee — accepted AI-conference papers

2026

Difficulty-Aware Label-Guided Denoising for Monocular 3D Object Detection

AAAI 2026technical

Monocular 3D object detection is a cost-effective solution for applications like autonomous driving and robotics, but remains fundamentally ill-posed due to inherently ambiguous depth cues. Recent DETR-based methods attempt to mitigate this through global attention and auxiliary depth prediction, ye

Cited by 0 Source ↗PDF ↗Scholar ↗

2025

TADFormer: Task-Adaptive Dynamic TransFormer for Efficient Multi-Task Learning

Seungmin Baek, Soyul Lee, Hayeon Jo, Hyesong Choi, Dongbo Min

CVPR 2025poster

Transfer learning paradigm has driven substantial advancements in various vision tasks. However, as state-of-the-art models continue to grow, classical full fine-tuning often becomes computationally impractical, particularly in multi-task learning (MTL) setup where training complexity increases prop…

Cited by 0 Source ↗PDF ↗Scholar ↗