Your Photo

Hello, I'm Daehyun Kim

I recently received my Ph.D. under the supervision of Prof. Tae Hyun Kim. My research focuses on low-level computer vision problems such as low-light image enhancement, exposure correction, and under-display camera imaging. I am also interested in exploring how these techniques can be applied to real-world visual perception and interaction problems. If you are interested in collaboration, feel free to contact me.

GitHub Google Scholar Email LinkedIn VILAB

Publications

UCMNet
UCMNet: Uncertainty-Aware Context Memory Network for Under-Display Camera Image Restoration
CVPR, 2026 | code
Daehyun Kim, Youngmin Kim, Yoon Ju Oh, Tae Hyun Kim
We propose a lightweight Uncertainty-aware Context-Memory Network (UCMNet), for UDC image restoration. Unlike previous methods that apply uniform restoration, UCMNet performs uncertainty-aware adaptive processing to restore high-frequency details in regions with varying degradations.
Exposure-slot
Exposure-slot: Exposure-centric representations learning with Slot-in-Slot Attention
CVPR, 2025 | code
Donggoo Jung*, Daehyun Kim*, Tae Hyun Kim
We propose Exposure-slot, a novel framework that integrates a prompt-based slot-in-slot attention mechanism to cluster exposed feature regions and learn exposure-centric features.
CLODE
CLODE: Continuous Exposure Learning for Low-Light Image Enhancement using Neural ODEs
ICLR, 2025 (spotlight) | code
Donggoo Jung*, Daehyun Kim*, Tae Hyun Kim
We formulate low-light enhancement as a continuous system using Neural ODEs, allowing stable curve-based exposure correction without sensitivity to iteration count.
SANFlow
SANFlow: Semantic-Aware Normalizing Flow for Anomaly Detection
NeurIPS, 2023 | code
Daehyun Kim, Sungyong Baik, Tae Hyun Kim
SANFlow maps features to diverse distributions based on semantic location, improving discrimination between normal and abnormal patterns.
Machine learning-based analysis for prediction of surgical necrotizing enterocolitis in very low birth weight infants using perinatal factors: a nationwide cohort study
European Journal of Pediatrics, 2024
Seung Hyun Kim, Yoon Ju Oh, Joonhyuk Son, Donggoo Jung, Daehyun Kim, Soo Rack Ryu, Jae Yoon Na, Jae Kyoon Hwang, Tae Hyun Kim, Hyun-Kyung Park
Learning-Based Longitudinal Prediction Models for Mortality Risk in Very-Low-Birth-Weight Infants: A Nationwide Cohort Study
Neonatology, 2023
Jae Yoon Na*, Donggoo Jung*, Jong Ho Cha, Daehyun Kim, Joonhyuk Son, Jae Kyoon Hwang, Tae Hyun Kim, Hyun-Kyung Park
Two-stage learning-based prediction of bronchopulmonary dysplasia in very low birth weight infants: a nationwide cohort study
Frontiers in Pediatrics, 2023
Jae Kyoon Hwang*, Dae Hyun Kim*, Jae Yoon Na*, Joonhyuk Son, Yoon Ju Oh, Donggoo Jung, Chang-Ryul Kim, Tae Hyun Kim, Hyun-Kyung Park
Development of artificial neural networks for early prediction of intestinal perforation in preterm infants
Scientific Reports, 2022
Joonhyuk Son*, Daehyun Kim*, Jae Yoon Na*, Donggoo Jung, Ja-Hye Ahn, Tae Hyun Kim, Hyun-Kyung Park

* indicates equal contribution

Academic Service