Wan-Cyuan Fan
Scholar

Wan-Cyuan Fan

Google Scholar ID: EIPHoLEAAAAJ
The University of British Columbia
Computer visionvision and language learningmulti-modal learning
Citations & Impact
All-time
Citations
234
 
H-index
7
 
i10-index
5
 
Publications
17
 
Co-authors
19
list available
Resume (English only)
Academic Achievements
  • Paper 'On Pre-training of Multimodal Language Models Customized for Chart Understanding' accepted by NeurIPS Workshop on Adaptive Foundation Models (Oct 2024)
  • Paper 'Surprising Observations in Basic Vision Language Model Capabilities' accepted by ECCVW (Aug 2024)
  • TAM-VT model ranked 5th in VOTS2024 Challenge at ECCV '24 (May 2024)
  • Paper 'IOU-Aware Multi-Expert Cascade Network via Dynamic Ensemble for Long-tailed Object Detection' accepted by ICASSP (Feb 2023)
  • Two papers accepted by AAAI (Nov 2022): 'Feature Pyramid Diffusion for Complex Scene Image Synthesis' and 'Target-free Text-guided Image Manipulation'
  • Paper 'Paraphrasing Is All You Need for Novel Object Captioning' accepted by NeurIPS 2022 (Sep 2022)
  • Master’s thesis on image manipulation awarded Honorable Master Thesis Award at IPPR 2022 (June 2022)
  • Paper 'Scene Graph Expansion for Semantics-Guided Image Outpainting' accepted by CVPR 2022 (Mar 2022)
  • Paper 'Cross-Modal Mutual Learning for Audio-Visual Speech Recognition and Manipulation' accepted by AAAI 2022 (oral presentation) (Dec 2021)
  • Paper 'LayoutTransformer: Scene Layout Generation with Conceptual and Spatial Diversity' accepted by CVPR 2021 (Feb 2021)
  • MEC detection model ranked 7th globally (Top 10) in the LVIS Challenge at ECCV workshop (Aug 2020)