Faculty

徐繼聖 Gee-Sern (Jison) Hsu photo
Position
Professor
Education
PhD in Mechanical Engineering from the University of Michigan, Ann Arbor
E-mail
jison@mail.ntust.edu.tw
Room
ME1-215

Office Hours
Telephone (02)2730-3234
Homepage https://sites.google.com/site/jisonhsuprofile2/
Laboratory 人工視覺實驗室
Major Field Generative AI, Computer vision, Defect detection, 3D perception
Course Generative Computer Vision (GCV)




Research Achievement
  [1]   Style-Preserving Generator for Synthetic License Plate Recognition (TCSVT 2025) https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11072464
  [2]   Automatic detect incorrect lifting posture with the pose estimation model (MDPI 2025) https://www.mdpi.com/2075-1729/15/3/358
  [3]   Multiview gait recognition on unconstrained path using graph convolutional neural network (IEEE 2022) https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9779706
  [4]   Masked face recognition from synthesis to reality (IEEE 2022) https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9738595
  [5]   Cross-spectrum thermal face pattern generator (IEEE 2022) https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9684443
  [6]   Motion Capture Sensor-Based Emotion Recognition Using a Bi-Modular Sequential Neural Network (MDPI 2022) https://www.mdpi.com/1424-8220/22/1/403
  [7]   Fast Landmark Localization With 3D Component Reconstruction and CNN for Cross-Pose Recognition (IEEE 2018) https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8023997
  [8]   Robust Cross-Pose Face Recognition Using Landmark Oriented Depth Warping (JVCIR 2018) https://www.sciencedirect.com/science/article/pii/S1047320318300695
  [9]   Hybrid Ageing Patterns for Face Age Estimation (Sciencedirect 2018) https://www.sciencedirect.com/science/article/pii/S026288561730121X
  [1]   Skeleton-Based Motion Pattern Recognition for Sarcopenia Risk Assessment
  [2]   AgeMoVLE: Age Estimation with Mixture of VL Experts (CVPRW 2026) https://openaccess.thecvf.com/content/CVPR2026W/FoundGen-Bio/papers/Chiu_AgeMoVLE_Age_Estimation_with_Mixture_of_VL_Experts_CVPRW_2026_paper.pdf
  [3]   Anchor-Guided Contrastive Learning for User Identification Based On Video Preferences
  [4]   Dynamic Age Estimation via Mixture of Experts: Bridging Semantic and Structural Models
  [5]   Attribute-Specified Generation And Style-Transfer Diffusion For Face Recognition Enhancement
  [6]   Style-Preserving Diffusion for Scene Text Editing
  [7]   Efficient Face Recognition via Representative Synthetic Data
  [8]   Syn3DTxt: Embedding 3D Cues for Scene Text Generation (CVPRW 2025) https://openaccess.thecvf.com/content/CVPR2025W/SyntaGen/papers/Hsiung_Syn3DTxt_Embedding_3D_Cues_for_Scene_Text_Generation_CVPRW_2025_paper.pdf
  [9]   Contour-Guided Context Learning for Scene Text Recognition
  [10]   A novel Laguerre Voronoi diagram token filtering strategy for computer vision transformers
  [11]   Pose Adapted Shape Learning for Large-Pose Face Reenactment
  [12]   Unleashing Stable Diffusion for Multi-Modal Facial Image Augmentation
  [13]   Outline Generation Transformer for Bilingual Scene Text Recognition
  [14]   Pretrained Pixel-Aligned Reference Network for 3D Human Reconstruction
  [15]   Traffic Violation Detection via Depth and Gradient Angle Change
  [16]   AgeTransGAN for facial age transformation with rectified performance metrics
  [17]   Graph Refinement with Regression Prior for 3D Face Reconstruction
  [18]   Dual-generator face reenactment
  [19]   Thailand license plate detection and recognition
  [20]   Age-style and alignment augmentation for facial age estimation
  [21]   Pose-guided and style-transferred face reenactment
  [22]   Deep Hybrid Network for Automatic Quantitative Analysis of Facial Paralysis
  [23]   Edge-Coupled and Multi-Dropout Face Alignment
  [24]   CoVieW'18: The 1st Workshop and Challenge on Comprehensive Video Understanding in the Wild
  [25]   Hierarchical Network for Facial Palsy Detection

Research Achievement
Title Total Amount Conduct From Conduct Until Client
基於視覺自回歸模型之主角驅動個人化生成與姿態-年齡解纏研究 3,750,000元 11508 11807 NSTC
Large-Pose and Cross-Age Face Reenactment 2,805,000元 11208 11507 NSTC

Student
Academic Year Thesis Title Student
M.S
Academic Year Thesis Title Student
Academic Year Project Title Student
Year Deeds