About me
I am currently a senior applied scientist at Amazon. My research area is in computer vision and generative AI for fashion. Before joining Amazon, I was a computer vision scientist at GE Global Research and a Postdoc at NVIDIA-NTU AI Lab. I completed my PhD in Computer Science at National Taiwan University under the supervision of Prof. Winston Hsu. I was a visiting researcher at University of Maryland working with Prof. Larry S. Davis.
Working Experience
Senior Applied Scientist, Amazon, California, USA (2019.08 to now)
Computer Vision Scientist, GE Global Research, New York, USA (2016.12 to 2019.06)
Postdoctoral Researcher, NVIDIA-NTU AI Lab, Taipei, Taiwan (2015)
Visiting Scholar, University of Maryland, College Park, MD, USA (2014)
Publications
Outfit Transformer: Learning Outfit-Level Representations for Fashion Recommendation
Rohan Sarkar, Navaneeth Bodla, Mariya I. Vasileva, Yen-Liang Lin, Anurag Beniwal, Alan Lu, Gerard Medioni
WACV 2023 [pdf]
SLADE: A Self-Training Framework for Distance Metric Learning
Jiali Duan, Yen-Liang Lin, Son Tran, Larry S. Davis, C.-C. Jay Kuo
CVPR 2021 [pdf]
Fashion Outfit Complementary Item Retrieval
Yen-Liang Lin, Son Tran, Larry S. Davis
CVPR 2020 [pdf]
A Unified Point-Based Framework for 3D Segmentation
Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, Winston Hsu
3DV 2019 [pdf], CVPR workshop 2019 (2nd place at ScanNet challenge)
DCAN: Dual Channel-Wise Alignment Networks for Unsupervised Scene Adaptation
Zuxuan Wu, Xintong Han, Yen-Liang Lin, Mustafa Gkhan Uzunbas, Tom Goldstein, Ser Nam Lim, Larry S. Davis.
ECCV 2018 [pdf]
Drone-based Object Counting by Spatially Regularized Regional Proposal Network
Meng-Ju Hsieh, Yen-Liang Lin, Winston Hsu
ICCV 2017 [pdf]
Joint Sequence Learning and Cross-Modality Convolution for 3D Biomedical Segmentation
Kuan-Lun Tseng, Yen-Liang Lin, Winston Hus, Chung-Yang Huang
CVPR 2017 [pdf]
Scalable Object Detection by Filter Compression with Regularized Sparse Coding
Ting-Hsuan Chao, Yen-Liang Lin, Yin-Hsi Kuo, Winston Hsu
CVPR 2015 [pdf]
Summarizing While Recording: Context-Based Highlight Detection for Egocentric Videos
Yen-Liang Lin, Vlad I. Morariu, Winston Hsu
ICCV workshop 2015 [pdf]
3D Representation for Recognition and Retrieval
Yen-Liang Lin
CVPR 2015 (Doctoral consortium)
Jointly Optimizing 3D Model Fitting and Fine-Grained Classification
Yen-Liang Lin, Vlad I. Morariu, Winston Hsu, Larry S. Davis
ECCV 2014 [pdf][code & dataset]
3D Sub-Query Expansion for Improving Sketch-Based Multi-View Image Retrieval
Yen-Liang Lin, Cheng-Yu Huang, Hao-Jeng Wang, Winston Hsu
ICCV 2013 [pdf]
Investigating 3D Model and Part Information for Improving Content-Based Vehicle Retrieval
Yen-Liang Lin, Ming-Kuang Tsai, Winston Hsu, Chih-Wei Chen
IEEE TCSVT 2013
Sketch-Based Image Retrieval on Mobile Devices Using Compact Hash Bits
Kai-Yu Tseng, Yen-Liang Lin, Chen-Yu Hsiu, Winston H. Hsu
ACM MM 2012 [pdf]
Snap2read: Automatic Magazine Capturing and Analysis for Adaptive Mobile Reading
Yu-Ming Hsu, Yen-Liang Lin, Chen-Yu Hsiu, Winston H. Hsu
MMM 2011 (Best paper award)
US Patent
11645693 - Complementary consumer item selection (2023/5/9)
US20210103726A1 - Building footprint (2021/04/08)
10475174 - Visual anomaly detection system (2019/11/12)
US Government Projects
CORE3D (IARPA) - collaboration with UMD, CMU, Cornel, TU Graz
The goal of the CORE3D program is to develop technology that automatically generates accurate 3D building models using multiple data sources from satellite imagery