Dong-Hwan Jang

I am a master's student at Seoul National University, Electrical and Computer Science Department, advised by Bohyung Han. I am currently at CMU as a visiting scholar, fully funded by Korea Government. I completed my Bachelors degree with Electrical and Computer Engineering & Technology Management from Seoul National University in 2020.

Email  /  CV  /  Google Scholar

profile photo
Research

I envision a machine that can dynamically adapt and evolve to the changing environment similar to our brain.

I believe current static neural networks call for the scale-up of their model size to compensate for the lack of adaptability, and make the models hard to be deployed in changing unstructured environments.
To address this issue, I am interested in dynamic neural networks that can make machines flexibly learn where to focus and evolve accordingly, like human brain.
My research interests include computer vision, machine learning, robotics, and neuroscience.

Pooling Revisited: Your Receptive Field is Suboptimal
Dong-Hwan Jang, Sanghyeok Chu, Joonhyuk Kim, Bohyung Han
CVPR, 2022

We propose DynOPool, a learnable resizing module that finds the optimal scale factors and receptive fields of intermediate feature maps.

DS4C Patient Policy Province Dataset: a Comprehensive COVID-19 Dataset for Causal and Epidemiological Analysis
Jimi Kim*, Seojin Jang*, Joong Kun Lee*, Dong-Hwan Jang* (* indicates equal contributions)
NeurIPS Workshop (Causal Discovery & Causality-Inspired Machine Learning), 2020

project page / article (korean) (auto-translated)

We present DS4C South Korea Patient, Policy, and Provincial data (DS4C-PPP dataset). The dataset contains comprehensive data that could be used for causal analysis, such as per-patient symptom onset and confirmed date, travel frequency, hospital accessibility, and 61 preventative policies enacted in South Korea.

Academic Projects
DIFF: Deblurring Implicit Feature Function

We propose spatially-variant motion deblur network based on the implicit neural representation. A spatially-variant deblurring network takes deformed features and their offsets as inputs.

U.S. Patent Application Number: 17/973,809 (in progress)
Dynamic Spatially-Adaptive Modulation for image Dehazing

We propose a novel framework of the dynamic spatially-adaptive modulation for image dehazing. The proposed algorithm introduces a selection module that conditionally determine the necessary modulation pathways in a bottom-up manner by providing a loss function optimizing both accuracy and efficiency.

DepthFinder: Universal Image Restoration based on Adaptive Inference

We adaptively find the appropriate number of the residual blocks according to the severity and distortion type of the input in universal image restoration task.


Talks
Korean Conference on Computer Vision 2022

20 minutes oral presentation (top 23.5% among published papers) on CVPR paper “Pooling Revisited: Your Receptive Field is Suboptimal” presented by prof. Bohyung Han

ds4c Databricks Invited Talk

1 hour talk on “The Complexities around COVID-19 Data” invited as DS4C team
Scholarships & Award

Korean Government Scholarship for Overseas Study (2023-2024)
 - Covers USD 40,000 support per year. Only 64 students are selected in all fields in Korea.

Hyundai OnDream Global Scholarship Award (2022)
 - Award Prize - around USD 2,350
 - For the paper “Pooling Revisited: Your Receptive Field is Suboptimal” at CVPR 2022

Hyundai OnDream Future Technology Scholarship (2021-2022)
 - Covers full tuition & financial support
Teaching Experiences

• Teaching Assistant for 430.329: Introduction to Algorithms at Seoul National University (Fall 2020)
• Teaching Assistant for Samsung AI Expert Course at Seoul National University (July 2019)
• Teaching Assistant for Hyundai Motors AI Expert Course at Seoul National University (Jan 2019)

Template: Jon Barron, Last update: 12/2022.