Hi, I'm Junsu 👋

Undergraduate @ KAIST (EE). Interests in Computer Vision & 3D Vision, multi-view reconstruction, and 3D Gaussian Splatting.

Junsu Kim

Education

Korea Advanced Institute of Science & Technology (KAIST)
Daejeon, South Korea
  • B.S. in Electrical Engineering
  • Mar 2021 – Expected Aug 2027
  • GPA: 4.10/4.3 (Major 4.20/4.3)

Honors & Activities

Honors

  • KAIST Presidential Fellowship (15th) Feb 2025 – Present
  • KAIST EE, Dean’s List Award Spring 2024 Awarded to the top 3% of all EE students
  • KAIST, Freshman Dean’s List Award Fall 2021 Awarded to the top 3% of all freshman students
  • Woon Hae Scholarship (12nd) Feb 2025 – Dec 2025

Activities

  • YEHS (Young Engineers Honor Society) Jan 2025 – Present
  • Include (KAIST AI Studying Club) Sep 2024 – Present
  • SilverLining (KAIST International Volunteer Club) Mar 2025 – Present

Skills

C/C++, Python, PyTorch

Publications

— (Add here when available)

Projects

Face Verification with AM-Softmax (EE40034, KAIST)

Course project on deep metric learning for face verification using ResNet-18 and AM-Softmax.

  • Built a two-stage training pipeline (Train1 from scratch, Train2 fine-tuning).
  • Performed ablation on loss, optimizer, scheduler, and augmentation strategies.
Computer Vision Metric Learning Face Verification

Experience

CVLAB, KAISTJun 2025 – Present

Research Intern (Advisor: Prof. Seungryong Kim, KAIST AI)

  • 3D Vision: multi-view reconstruction, 3D Gaussian Splatting
VICLAB, KAISTDec 2024 – Jun 2025

Research Intern (Advisor: Prof. Mun Churl Kim, KAIST EE)

  • Reviewed RCNN–YOLO, ViT and implemented few-shot/anomaly detection baselines
© 2025 Junsu Kim · Contact