More Realistic Than Reality!

  • home
  • 교수소개
  • 전임교수



  • 조교수 컴퓨터비전, 멀티모달학습, 전이학습
  • 홍성은 홈페이지 바로가기
    Lab 인공지능 및 미디어 랩


다중 모달 학습, 시각-언어 모델, 도메인 적응, 로봇 비전


  • Ph.D., Computer Science, KAIST (2018)
  • MS, Computer Science, KAIST (2012)
  • BS, Computer Engineering, Hanyang University (2010)


  • 조교수, 인하대학교 정보통신공학과 (2020 - 2023)
  • Research Scientist, SK텔레콤 T-Brain (2018 - 2020)

학술지 논문

  • (2023)  TL-ADA: Transferable Loss-based Active Domain Adaptation.  NEURAL NETWORKS.  161,  1
  • (2022)  Adaptive Graph Adversarial Networks for Partial Domain Adaptation.  IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY.  32,  1
  • (2021)  Domain Adaptation Without Source Data.  IEEE Transactions on Artificial Intelligence.  2,  6
  • (2021)  Self-Supervised Feature Enhancement Networks for Small Object Detection in Noisy Images.  IEEE SIGNAL PROCESSING LETTERS.  28,  1
  • (2020)  Towards Privacy-Preserving Domain Adaptation.  IEEE SIGNAL PROCESSING LETTERS.  27,  1
  • (2020)  Attention-guided adaptation factors for unsupervised facial domain adaptation.  ELECTRONICS LETTERS.  56,  16
  • (2019)  Unsupervised Face Domain Transfer for Low-Resolution Face Recognition.  IEEE SIGNAL PROCESSING LETTERS.  27,  1
  • (2018)  D3: Recognizing dynamic scenes with deep dual descriptor based on key frames and key segments.  NEUROCOMPUTING.  273,  1
  • (2018)  Not all frames are equal: aggregating salient features for dynamic texture classification.  MULTIDIMENSIONAL SYSTEMS AND SIGNAL PROCESSING.  29,  1
  • (2015)  Sorted Consecutive Local Binary Pattern for Texture Classification.  IEEE TRANSACTIONS ON IMAGE PROCESSING.  24,  1
  • (2013)  Recursive Bayesian fire recognition using greedy margin-maximizing clustering.  MACHINE VISION AND APPLICATIONS.  24,  1


  • Method and Apparatus for Transferable Loss_based Active Domain Adaptation.  C-2023-020239.  20230504.  대한민국
  • 군중 계산을 위한 교차 모달 공간 채널 주의 블록.  C-2022-047209.  20221122.  대한민국
  • 촬영 환경에 강인한 차량 번호판 인식 시스템.  C-2012-021243.  20120604.  대한민국


  • 한국 MEMS 학술대회 최우수 논문상, 2023.03
  • 한국방송·미디어공학회 신진연구자상, 2021.12
  • Vision Meets Drones 2019: A Challenge (ICCV '19) 세계 3위 입상, 2019.10
  • Unsupervised DAVIS Challenge on Video Object Segmentation (CVPR '19) 세계 3위 입상, 2019.06
  • SK텔레콤 Tech Expert Board (TEB) 대상 수상, 2018.10
  • VizWiz Grand Challenge: Answering Visual Questions from Blind People (ECCV '18) 세계 2위 입상, 2018.09


  • (2023)  Intra-Inter Modal Attention Blocks for RGB-D Semantic Segmentation.  ACM International Conference on Multimedia Retrieval (Former ACM International Conference on Image and Video Retrieval).  그리스
  • (2023)  IFQA: Interpretable Face Quality Assessment.  Winter Conference on Applications of Computer Vision.  미국
  • (2022)  Spatio-channel Attention Blocks for Cross-modal Crowd Counting.  Asian Conference on Computer Vision.  중국
  • (2021)  Patch-level Augmentation for Object Detection in Aerial Images.  Workshops at International Conference on Computer Vision.  대한민국
  • (2021)  Re-Aging GAN: Toward Personalized Face Age Transformation.  International Conference on Computer Vision.  미국
  • (2019)  Key Instance Selection for Unsupervised Video Object Segmentation.  Workshops at Computer Vision and Pattern Recognition.  미국
  • (2018)  Scale-Varying Triplet Ranking with Classification Loss for Facial Age Estimation.  Asian Conference on Computer Vision.  오스트레일리아
  • (2018)  Bilinear Attention Networks for VizWiz Challenge.  Workshops at European Conference on Computer Vision.  독일
  • (2018)  Cascaded Pyramid Network for 3D Human Pose Estimation Challenge.  Workshops at European Conference on Computer Vision.  독일
  • (2018)  CBVMR: Content-Based Video–Music Retrieval Using Soft Intra-Modal Structure Constraint.  International Conference on Multimedia Retrieval.  일본
  • (2017)  SSPP-DAN: Deep Domain Adaptation Network for Face Recognition with Single Sample Per Person.  International Conference on Image Processing.  중국
  • (2013)  3D Keypoint Tracking Based on Hybrid Flow Computation for Human Motion Analysis.  International Conference on Machine Vision Applications.  일본