News 

• [07/2024] One paper is accepted to ECCV.

• [04/2024] Two papers are accepted to CVPR Workshop on Efficient and On-Device Generation (EDGE).

• [03/2024] One paper is accepted to ICLR Workshop on Mathematical & Empirical Understanding of Foundation Models (ME-FoMo).

• [06/2023] One paper is accepted to ICML Workshop on Efficient Systems for Foundation Models (ES-FoMo).

• [05/2023] Two demos are accepted to ICCV Demo Track (available at Hugging Face Spaces).

• [04/2023] One paper is accepted to MLSys Workshop on On-Device Intelligence (ODIW).

• [01/2023] Our work is accepted to NVIDIA GPU Technology Conference (GTC) as Poster Spotlight.

• [12/2022] Our work on supporting SegFormer is integrated into NetsPresso 1.2.0.

• [10/2022] One paper is accepted to IEEE Transactions on Instrumentation & Measurement (TIM).

• [06/2022] One paper is accepted to ICML Workshop on Hardware Aware Efficient Training (HAET).

• [03/2022] Our work is accepted to NVIDIA Applied Research Accelerator Program (ARAP).

• [11/2021] I'm thrilled to join Nota Inc.

Research Interest — P: preprint, J: journal, C: conference/workshop, A: abstract 

I am currently interested in neural network compression (e.g., pruning [C15, C14, C12, P3, C11, C10, C9], knowledge distillation [C15, C14, C13, C11, C10], quantization [C13, C10]).

My research interests are in the areas of machine learning to process:

• Image/Video: text-to-image generation [C15, C14, C13, C11], talking face generation [C10, C9], RGB-NIR video fusion [J8], controllable image synthesis [C9, Thesis, P1, J6], fashion image manipulation [J6, A8], facial expression recognition [J4, C7, C6, A7, Challenge Winner], object recognition [C4]

• Text: language modeling [C12, P3, J7], dialogue act classification [C8], document classification [J1]

• Biosignal: rPPG-based heart rate estimation [J8], EEG classification [J3, C2, A1] & analysis [C5, A6, A4], fMRI analysis [J2]

• Speech: speech enhancement [J5], emotional sound recognition [C3], intention recognition [A3]

Publication

Please refer to the publication page.

Education

• Ph.D., Electrical Engineering (Machine Learning), KAIST, 2020

  • Thesis: Learning-Based Image Synthesis with Disentangled Representations [PhD Defense] [Grad Research Portfolio]

  • Advisor: Soo-Young Lee, GPA: 4.24 / 4.3, Integrated MS-PhD Program

• B.S., Bio and Brain Engineering, KAIST, 2012

  • Ranked 5th among all undergraduates at 2013 Commencement (5/838 = 0.6%)

  • GPA: 4.11 / 4.3 (summa cum laude)

Employment

• Senior Researcher at Nota Inc. — Nov. 2021–Present

• Senior Researcher at Samsung SDS — Apr. 2021–Oct. 2021

• Postdoctoral Researcher at KAIST (Prof. Eunho Yang’s Group) — Apr. 2020–Aug. 2020

• Undergraduate Researcher at KAIST (Prof. Soo-Young Lee’s Group) — Dec. 2011–Aug. 2012

• Research Intern at KRISS — June 2011–Aug. 2011

Honor

• Applied Research Accelerator Program (ARAP), NVIDIA, 2022

• Excellence Award, Annual PhD Research Progress Evaluation, School of EE, KAIST, 2017

• Challenge Winner, The 3rd Emotion Recognition in the Wild Challenge (EmotiW’15), ICMI, 2015

• Qualcomm Innovation Award, Qualcomm Korea Corp. and KAIST, 2013

• Graduate Full Scholarship from Korean Government, 2012–2019

• Ranked 5th among All 838 Undergraduates based on GPA, KAIST Commencement, 2013

• Undergraduate Full Scholarship from STX Scholarship Foundation, 2009–2012

Academic Service

• Conference/Workshop Reviewing

  • Neural Information Processing Systems (NeurIPS) —  2024∙2023∙2022∙2021∙2020

  • International Conference on Machine Learning (ICML) — 2024∙2023∙2022

  • CVPR Workshop on Efficient Large Vision Models (ELVM) — 2024

  • IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) — 2024

  • International Conference on Learning Representations (ICLR) — 2024∙2023∙2022

  • ECCV Workshop on Computational Aspects of Deep Learning (CADL) — 2022

  • ACM International Conference on Multimodal Interaction (ICMI) — 2017

• Journal Reviewing

  • Neurocomputing, Elsevier — 2022∙2018

  • Pattern Recognition, Elsevier — 2021∙2020∙2017

  • IEEE Transactions on Affective Computing — 2020

  • IEEE Transactions on Multimedia — 2019

  • IEEE Access — 2019

  • Neural Processing Letters, Springer — 2015

Talks

• “Style-Controlled Synthesis of Clothing Segments for Fashion Image Manipulation,” Kyungpook National Univ., Daegu, Korea, 2019

• “AI for Fashion,” Sookmyung Women's Univ., Seoul, Korea, 2019

• “AI for Fashion,” ETRI, Daejeon, Korea, 2017

• “Fusing Aligned and Non-Aligned Face Information for Automatic Affect Recognition in the Wild,” CVPRW, Las Vegas, NV, 2016

• “Hierarchical Committee of Deep CNNs for Static Facial Expression Recognition,” ICMI, Seattle, WA, 2015

• “Spectral Feature Extraction Using dNMF for Emotion Recognition in Vowel Sounds,” ICONIP, Daegu, Korea, 2013