We accelerate vision and learning algorithms π
We accelerate vision and learning algorithms π
π‘ Research Topics
We develop AI-based Intelligent Surveillance, leveraging video and sensor data, based on various tasks including object detection, tracking, anomaly detection, and simulation aiming to build secure and reliable reconnaissance systems.
We focus on Perception, Reasoning, and Action based on multi-modal data such as images across various spectrums, and text, developing next-generation AI that enables complex situation understanding and intelligent decision-making.
We concentrate on Representation Learning, enabling efficient understanding and generalization of high-dimensional data, thereby improving performance and efficiency across diverse AI applications.
π£ Latest News
κΉμμ°¬ λ°μ¬ νμλ Όλ¬Έλͺ : Learning for Visual Understanding: A Data-Centric Perspective (Ph.D Thesis)
μΆνν©λλ€. G-MAD: A Game-Based Data Generation Framework for Multi-View RGB-T Aerial Object Detection (Yechan Kim et al.) μ£Όμ μ μκ³ λ 4λͺ μ νΌμ΄ 리뷰λ₯Ό κ±°μΉ λ€, νλ‘κ·Έλ¨ μμνκ° μ΅μ’ μλ½νμΌλ©°, ACM MM 2026 λ©μΈ νλ‘μλ©μ μ€λ¦΄ μμ μ λλ€. (* μ¬ν΄ 11μ λΈλΌμ§μμ ORAL λ°ν μμ *) ν΄λΉ λ Όλ¬Έμ LIG D&Aμ ν¨κ» μνν μ°κ΅¬ κ²°κ³Όμ λλ€.
κΉμμ°¬ (GIST μ μ»΄ λ°μ¬) μ°κ΅¬μμ 26λ 6μ 1μΌμλ‘ LIG D&Aμ μ‘Έμ μ μ‘°κΈ° μμ (μ°νμ₯νμμΌλ‘ μ λ°λ ν μμ μ±μ©μΌλ‘ μ ν)λμμ΅λλ€.Β κΉμμ°¬ λ°μ¬λ λͺλͺ κ΅κ° μ°κ΅¬κΈ°κ΄μλ 1μ°¨ ν©κ²©ν μνμμΌλ, λ°©μ° μΈκ³΅μ§λ₯μ΄λΌλ μ½μ§ μμ νμ€ν¬μ λμ νκΈ° μν΄ LIG D&A AIλ© (νκ΅νμ°μ€)μ μ΅μ’ μ ννμμ΅λλ€. κΉμμ°¬ λ°μ¬λ LIG D&A κΈ°μ νμ λ³ΈλΆ μ°ν AIλ© (κ°μ λΉ μν λΆμ)μμ νλν°μ΄ ν ν¬λλ‘μ§ λ± μ κ΄ μ 체μ ν¨κ» νλ ₯νμ¬ μ μ₯ μν© μΈμ λ° λ°©μ± μμ±κ³Ό κ΄λ ¨ν μΈκ³΅μ§λ₯ μμ² κΈ°μ μ μ°κ΅¬ κ°λ°ν μμ μ λλ€.Β
λ°μ’ ν μ€λ¬΄μ± μμ λ° κΉμμ°¬ μ°κ΅¬μμ΄ νμ΄ NRF ν΅μ¬μ°κ΅¬ μμ΄λμ΄κ° NRFμ μν΄ μ μ λμ΄ νλλ₯Ό μ§μ λ°κ² λμμ΅λλ€. μΆνν©λλ€.
κ³Όμ λͺ : μλ νμ΄λ°μ΄μ λͺ¨λΈ κΈ°λ°μ μΈκ°-AI μνΈμμ©ν μμ§ν μκ°μ§λ₯ μμ΄μ νΈ μ°κ΅¬Β·κ°λ° (3λ ν)
κΉμμ° (GIST μ μ»΄ μμ¬) μ°κ΅¬μμ΄ 26λ 1μ 19μΌμλ‘ KRIT μ κ·μ§ μ°κ΅¬μμΌλ‘ μμ©λμμ΅λλ€.
κΉμμ°¬ (GIST μ μ»΄ λ°μ¬κ³Όμ ) μ°κ΅¬μμ μμμ μΆνν©λλ€.Β
μ£Όμ’ λ―Ό (GIST μ μ»΄ νμ¬κ³Όμ ) μΈν΄ μ°κ΅¬μμ μμμ μΆνν©λλ€.
λ§λ°λ¦° λ°μ¬ νμλ Όλ¬Έλͺ : Visual Multi-Object Tracking with Re-Identification using Labeled Random Finite Sets (Ph.D Thesis)
κ³ μλ―Ό λ°μ¬νμ°κ΅¬μ (GIST μ μ»΄ ν/μλ°)μ΄ 24λ 9μ 1μΌμλ‘ κ²½λΆλ λ‘λ΄κ³΅νκ³Ό μ‘°κ΅μ (RAG Lab)λ‘ μμ©λμμ΅λλ€. μΆνν©λλ€. πΒ
πΌοΈ Gallery
π Contact
π
Please feel free to contact our research team.
π¨βπ« Prof. Moongu Jeon
mgjeon@gist.ac.kr
+82-62-715-2406 / EECS-C3-308
π
We are always looking for strong, motivated students.
π¨βπ GIST E-mentor Jonghyun Park
citizen135@gm.gist.ac.kr
+82-62-715-3254 / EECS-C3-306