Welcome to my Portfolio!

So Glad to Meet You.

Take a Tour of my Work

About Me

Ph.D. student in Computer Vision

I am Dayoung, a Ph.D. student at Soongsil University’s Graduate School, under the guidance of Prof. Seong Heum Kim in the Visual Intelligence and Platform (VIP) Lab. My research focuses on hardware-efficient neural network architectures, encompassing Efficient Deep Learning, Pruning, Vision Transformers, Hybrid Architectures, Image Classification, and Object Detection. Recently, I have been particularly exploring Hybrid CNN-Transformer architectures, Pruning strategies, and advancements in Efficient Deep Learning.

Publications

Our research publications

One-cycle Structured Pruning with Stability Driven Structure Search

One-cycle Structured Pruning with Stability Driven Structure Search

Authors: Deepak Ghimire, Dayoung Kil, Seonghwan Jeong, Jaesik Park, Seong-heum Kim

Publication: arXiv, 2024


A Study of Structure Pruning for Hybrid Neural Networks

A Study of Structure Pruning for Hybrid Neural Networks

Authors: Deepak Ghimire, Dayoung Kil and Seong-heum Kim

Publication: International Conference on Control Automation and Systems, 2024


Lightweight Room Layout Estimation using a Single Panoramic Image

Lightweight Room Layout Estimation using a Single Panoramic Image

Authors: Dayoung Kil, Seong-heum Kim

Publication: International Conference on Control Automation and Systems, 2022


Lightweight Deep Learning for Room Layout Estimation with a Single Panoramic Image

Lightweight Deep Learning for Room Layout Estimation with a Single Panoramic Image

Authors: Dayoung Kil, Seong-heum Kim

Publication: Journal of Institute of Control, Robotics and Systems (SCOPUS), 2022


A Survey on Efficient Convolutional Neural Networks and Hardware Acceleration

A Survey on Efficient Convolutional Neural Networks and Hardware Acceleration

Authors: Deepak Ghimire, Dayoung Kil and Seong-heum Kim

Publication: Electronics (SCIE) 2022


Projects

Explore the work I’ve been involved in.

Zero-Shot Fall Detection FrameWork
Zero-Shot Fall Detection: Leveraging Text Prompt in Vision-Language Models

2023.08 - 2023.11

Multi-modal AI-based Firtual Fitting APP Service FrameWork
V Fit : Multi-modal AI-based Firtual Fitting APP Service

2023.06 - 2023.09

Food Calorie Estimation FrameWork
Food Calorie Estimation

2022.04 - 2022.06

A Website that Mosaic People's Faces through Face Recognition FrameWork
Face Out : A Website that Mosaic People's Faces through Face Recognition

2021.03 - 2021.11

3D Room Reconstruction Using HorizonNet FrameWork
3D Room Reconstruction Using HorizonNet

2021.05 - 2021.06

Awards

Recognitions received

  • Nov 19, 2024

    최우수상 (ITTP 원장상)

    제 1회 자율주행 인공지능 챌린지

    주관/주최: 과학기술정보통신부 / ETRI, kakao mobility, IITP, KADIF, 자율주행 DNA 기술포럼

    수상 내용: 차량용 객체복합 상태 인식 - 주행환경의 차량/버스의 분류(Agent), 의미론적 위치(Location), 후미등 상태(State)를 인식하는 동시에 Instance Segmentation을 진행.

  • Nov 12, 2021

    우수논문상

    졸업논문발표회

    주관/주최: SoongSil University School of AI Convergence

    수상 내용: 단일 파노라마 입력의 실내 공간 레이아웃 복원 모델 경량화 (Lightweight Deep Learning for Room Layout Estimation with a Single Panoramic Image)

  • image alt text

    Nov 12, 2021

    장려상

    AI융합 경진대회

    주관/주최: SoongSil University School of AI Convergence

    수상 내용: Face Out - Face Recognition을 통해 지정된 사람을 제외한 다른 사람들의 얼굴을 모자이크 처리하는 웹 서비스

  • Sep 28, 2021

    장려상

    제 11회 숭실 캡스톤디자인 경진대회

    주관/주최: SoongSil University

    수상 내용: Face Out - Face Recognition을 통해 지정된 사람을 제외한 다른 사람들의 얼굴을 모자이크 처리하는 웹 서비스

  • Feb 8, 2021

    우수상

    숭실 A3+ STAR 코딩 경진대회

    주관/주최: AI 모빌리티사업단 of SoongSil University

  • More
    to
    come!

Education

My academic history includes the following:

Mar 2024 - Now

Doctor of Intelligent Semiconductors

Soongil University, Seoul, South Korea
Mar 2022 - Feb 2024

Master of Intelligent Systems

Soongil University, Seoul, South Korea
Mar 2018 - Feb 2022

Bachelor of AI Convergence

Soongil University, Seoul, South Korea