My name is Wan-Cyuan (Chris) Fan, and I am a Ph.D. student working with Prof. Leonid Sigal on the topics of cross-modal learning, especially on vision-language synthesis and editing, at the University of British Columbia.
I previously worked as a Research Assistant in the Vision and Learning Lab (VLL) at National Taiwan University (NTU), guided by Prof. Yu-Chiang Frank Wang during my Master's and Research Assistant journey. Additionally, I collaborated with Yen-Chun Chen, DongDong Chen, Yu Cheng, and Lu Yuan as a student intern at Microsoft Research for six months. I received my Bachelor of Science degree in Electrical Engineering from NTU in 2020, during which I served as a student research intern for a year at the Institute of Information Science, Academia Sinica, under the guidance of Prof. Tyng-Luh Liu.
My research interests include Deep Learning (DL) and Computer Vision (CV), with a focus on multi-modal learning, x-to-image synthesis, text-guided image manipulation and editing, video generation, and object detection/segmentation.
News Red: academic activity; Green: internship activity
Aug. 2023 - I completed my conscription as a rifleman serving in the attack helicopter group.
Feb. 2023 - Our paper of "IOU-Aware Multi-Expert Cascade Network via Dynamic Ensemble for Long-tailed Object Detection" are accepted by ICASSP.
Nov. 2022 - Our papers of "Feature Pyramid Diffusion for Complex Scene Image Synthesis" and "Target-free Text-guided Image Manipulation" are accepted by AAAI.
Oct. 2022 - I completed my first 40-min tech talk in Microsoft Research. Thank Yen-Chun Chen and XiYang Dai for giving such great opportunity!
Sep. 2022 - Our paper entitled "Paraphrasing Is All You Need for Novel Object Captioning" is accepted by NeurIPS 2022.
June 2022 - My master thesis about image manipulation is selected as the honorable master thesis award in IPPR 2022.
March 2022 - Our paper entitled "Scene Graph Expansion for Semantics-Guided Image Outpainting" is accepted by CVPR 2022.
March 2022 - I joined Microsoft Research as a student intern.
Jan. 2022 - I completed my master thesis defense presentation! Thank Prof. Yu-Chiang Frank Wang, Prof. Chu-Song Chen, and Prof. Wei-Chen Walon Chiu for advising.
Dec. 2021 - Our work entitled "Cross-Modal Mutual Learning for Audio-Visual Speech Recognition and Manipulation" is accepted by AAAI 2022 (oral).
Feb. 2021 - Our work entitled "LayoutTransformer: Scene Layout Generation with Conceptual and Spatial Diversity" is accepted by CVPR 2021.
Sep. 2020 - I started my graduate student life.
Aug. 2020 - Our MEC detection model achieves Top 10 Winner (World Ranking 7) in the Large Vocabulary Instance Segmentation (LVIS) Challenge, ECCV workshop.
Jane 2020 - Due to the outbreak of covid-19 pandemic in California, the internship in ICT was canceled ;(
March 2020 - I received my first abroad job as a summer intern at Institute for Creative Technologies, USC. Thank Dr. Andrew (Wei-Wen) Feng and Dr. Meida Chen for giving me such an opportunity!
July 2019 - I joined Computer Vision & Machine Learning Lab, advised by Prof. Tyng-Luh Liu, as an undergraduate student intern in IIS, Sinica, Taiwan.
Publications (please wait a moment for loading gif.)
M3T: Multi-Scale Memory Matching for Video Object Segmentation and Tracking
Raghav Goyal*, Wan-Cyuan Fan*, Mennatullah Siam, Leonid Sigal
Under submission
[Paper] [Project Page]
IOU-Aware Multi-Expert Cascade Network for Long-tailed Object Detection
Wan-Cyuan Fan*, Cheng-Yao Hong*, Yen-Chi Hsu, Tyng-Luh Liu
ICASSP, 2023
[Paper]
Frido: Feature Pyramid Diffusion for Complex Scene Image Synthesis
Wan-Cyuan Fan, Yen-Chun Chen, Dongdong Chen, Yu Cheng, Lu Yuan, Yu-Chiang Frank Wang
AAAI, 2023
[Paper] [Project page] [Code]
Target-free Text-guided Image Manipulation
Wan-Cyuan Fan, Cheng-Fu Yang, Chiao-An Yang, Yu-Chiang Frank Wang
AAAI, 2023
[Paper (ArXiv pre-print)] [Code (coming soon)]
Paraphrasing Is All You Need for Novel Object Captioning
Cheng-Fu Yang, Yao-Hung Hubert Tsai, Wan-Cyuan Fan, Yu-Chiang Frank Wang, Louis-Philippe Morency, Ruslan Salakhutdinov
Thirty-Sixth Annual Conference on Neural Information Processing Systems, 2022
[Paper]
Scene Graph Expansion for Semantics-Guided Image Outpainting
Chiao-An Yang, Cheng-Yo Tan, Wan-Cyuan Fan, Cheng-Fu Yang, Meng-Lin Wu, Yu-Chiang Frank Wang
IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2022
[Paper] [Code (coming soon)]
Multi-expert heads of Long-tailed Instance Segmentation
Wan-Cyuan Fan, Cheng-Yao Hong, Yen-Chi Hsu, Tynh-Luh Liu
Tech. Report of IEEE European Conference on Computer Vision Workshop (ECCVW), 2020
[Paper]
Pre-prints and Miscellaneous
ACE: Adaptive Confusion Energy for Natural World Data Distribution
Yen-Chi Hsu, Cheng-Yao Hong, Wan-Cyuan Fan, Ming-Sui Lee, Davi Geiger, Tyng-Luh Liu
under submission (Journal), 2021
Auto-drawer: Generating and Modifying Images Continually byVisual-Relational Knowledge Graph
Wan-Cyuan Fan
Submitted to Second Workshop on Computer Vision for Fashion, Art and Design, ICCVw 2019
[Paper]
Awards
UBC 4YF Fellowship
The University of British Columbia, Canada, 2023
only 2-3 students each year in the CS department
NSERC PGSD/CGSD Award
Natural Sciences and Engineering Research Council of Canada, 2023
Honorable Master Thesis Award
Chinese Image Processing and Pattern Recognition Society, Taiwan, 2022
only 11 recipients in Taiwan
Research Scholarship
Novatek Foundation, Taiwan, 2021
only 3 recipients in NTU EECS (500+ students)
7th place
Large Vocabulary Instance Segmentation Challenge, ECCV workshop, 2020
Top 1 among non-industrial teams
Professional Activities
Reviewer
AAAI 2021, CVPR 2021, ICCV 2021, CVPR 2022, AAAI 2022, NeurIPS 2022