Latest News

---

[07/2022] Our paper on weakly supervised object detection is accepted at ECCV 2022.

[04/2022] Our paper on semi-weakly supervised semantic segmentation is accepted at IJCAI 2022.

[03/2022] Our paper on scene graph generation is accepted at IEEE TNNLS.

[08/2021] I recieved LLNL Computational Engineering Division Recognition Award.

 

---

Education

---

Ph.D. Seoul National University

• Mar 2015 ~ Feb 2020
• Computer Science and Engineering
• Advisor: Gunhee Kim

• Thesis: Improving Object Detection in Hard Conditions of Scale, Occlusion and Label

M.S. Seoul National University

• Mar 2013 ~ Feb 2015
• Computer Science and Engineering
• Advisor: Robert Ian McKay

• Thesis: Machine Learning Models and Missing Data Imputation Methods in Predicting the Progression of IgA Nephropathy

B.S. Seoul National University

• Mar 2005 ~ Feb 2013
• Computer Science and Engineering
• Statistics (Double Major)

• Thesis: Prediction of Customer’s Follow-on Purchase using Ensemble Methods

---

Selected Papers (*: Equal Contribution)

---

• Jinhwan Seo, Wonho Bae, Danica Sutherland, Junhyug Noh*, and Daijin Kim*. Object Discovery via Contrastive Learning for Weakly Supervised Object Detection. ECCV 2022.

• Wonho Bae, Junhyug Noh, Milad Jalali Asadabadi, and Danica Sutherland. One Weird Trick to Improve Your Semi-Weakly Supervised Semantic Segmentation Model. IJCAI 2022.

• Sangmin Woo, Junhyug Noh, and Kangil Kim. Tackling the Challenges in Scene Graph Generation with Local-to-Global Interactions. IEEE TNNLS 2022.

• Junhyug Noh*, Wonho Bae*, and Gunhee Kim. Rethinking Class Activation Mapping for Weakly Supervised Object Localization. ECCV 2020.

• Junhyug Noh*, Kyung Don Yoo*, Wonho Bae, …, and Jung Pyo Lee. Prediction of the Mortality Risk in Peritoneal Dialysis Patients using Machine Learning Models: A Nation-wide Prospective Cohort in Korea. Scientific Reports 2020.

• Junhyug Noh, Wonho Bae, Wonhee Lee, Jinhwan Seo, and Gunhee Kim. Better to Follow, Follow to Be Better: Towards Precise Supervision of Feature Super-Resolution for Small Object Detection. ICCV 2019.

• Junhyug Noh, Soochan Lee, Beomsu Kim, and Gunhee Kim. Improving Occlusion and Hard Negative Handling for Single-Stage Pedestrian Detectors. CVPR 2018.

• Junhyug Noh*, Kyung Don Yoo*, Hajeong Lee, …, and Yon Su Kim. A Machine Learning Approach Using Survival Statistics to Predict Graft Survival in Kidney Transplant Recipients: A Multicenter Cohort Study. Scientific Reports 2017.