Wang Wei-Cheng
Wang Wei-Cheng 王偉丞

Ph.D.
Identifying where lab-validated ML breaks in the field.

Based in Ghent until August 2026, returning to Taiwan thereafter.
Open to AS / RE roles, remote interviews available now.

I am currently working on synthetic data generation, extending my PhD research line to cross-environment ML transfer and explainable AI. Recently built a cost-efficient multi-agent LLM orchestration pipeline as PoC, and submitted a Google TPU Research Cloud proposal.

In 2025, I completed my Ph.D. at Ghent University-imec, and was previously a RA at Academia Sinica. Both my PhD and MSc research were conducted on real-world data, covering 5000+ hours of uncurated audio streams and extended video across multiple deployment sites.

My research focuses on building solutions at the mechanism-level while under real-world constraints. Recognition includes 12 peer-reviewed articles (7 journals, ICIP oral), 140+ citations, APICTA award, and Top 3% master’s graduate honor.

Download CV
Interests
  • Deep Learning
  • Computer Vision
  • Audio Privacy
Education
  • PhD in Computer Science Engineering

    Ghent University, Belgium

  • MSc in Computer and Communication Engineering

    National Cheng Kung University, Taiwan

  • BSc in Electrical Engineering

    National Cheng Kung University, Taiwan

Featured Projects
Publications
(2025). Embedding-based pair generation for contrastive representation learning in audio-visual surveillance data. Frontiers in Robotics & AI(JCR Q2).
(2024). Privacy-preserving visual analysis: training video obfuscation models without sensitive labels. Applied Intelligence(JCR Q2).
(2022). An Opt-in Framework for Privacy Protection in Audio-Based Applications. IEEE Pervasive Computing.
(2022). Selective manipulation of disentangled representations for privacy-aware facial image processing. In MLCS @ ECML PKDD 2022.
(2019). Driver Monitoring Using Sparse Representation With Part-Based Temporal Face Descriptors. IEEE T-ITS.
(2018). Clustering Trajectories in Heterogeneous Representations for Video Event Detection. IEEE ICIP 2018.
(2018). USEAQ: Ultra-Fast Superpixel Extraction via Adaptive Sampling From Quantized Regions. IEEE T-IP.
(2018). Spatiotemporal Coherence-Based Annotation Placement for Surveillance Videos. IEEE T-CSVT.
(2017). Event based surveillance video synopsis using trajectory kinematics descriptors. IAPR MVA 2017.
(2015). Video gender recognition using temporal coherent face descriptor. IEEE/ACIS SNPD 2015.
(2015). Trajectory kinematics descriptor for trajectory clustering in surveillance videos. IEEE ISCAS 2015.
Awards


Best Master's Thesis Award - 2016
  The Chinese Image Processing and Pattern Society
  The Chinese Institute of Electrical Engineering
Honourary Member - 2016
  The Phi Tau Phi Scholastic Honour Society
Merit Award - 2015
  Asia Pacific ICT Alliance (APICTA) Awards, R&D category, representing Taiwan
1st place - 2015
  International ICT Innovative Services Awards
2nd place - 2015
  International ICT Innovative Services Awards