Kun-Yu Lin

I am now a post-doctoral research fellow at the University of Hong Kong, under the supervision of Prof. Kai Han. I obtained my PhD degree from Sun Yat-set University, under the supervision of Prof. Wei-Shi Zheng. Prior to this, I obtained my Bachelor's degree and Master degree from Sun Yat-Sen University. During my PhD, I was fortunate to have the opportunity to study as a visiting student at MMLab@NTU, under the supervision of Prof. Chen Change Loy and Prof. Henghui Ding. My research interests include computer vision and machine learning.

Email  /  Scholar  /  Github

profile photo

News

        ❅ 07/2024: One paper was accepted to TPAMI.
        ❅ 03/2024: Releasing XOV-Action, the first cross-domain open-vocabulary action recognition benchmark!
        ❅ 09/2023: One paper was accepted to NeurIPS2023.
        ❅ 09/2023: One paper was accepted to TPAMI.
        ❅ 07/2023: One paper was accepted to ICCV2023.
        ❅ 03/2023: Two papers were accepted to CVPR2023.
 

Selected Works

Most of my research works are about human video understanding, transferable, generalizable and trustworthy deep learning, and robot learning. Some works are highlighted. # denotes equal contributions.

Rethinking CLIP-based Video Learners in Cross-Domain Open-Vocabulary Action Recognition
Kun-Yu Lin, Henghui Ding, Jiaming Zhou, Yu-Ming Tang, Yi-Xing Peng, Zhilin Zhao, Chen Change Loy, Wei-Shi Zheng
arXiv, 2024
arXiv / github

The first benchmark, named XOV-Action, for the cross-domain open-vocabulary action recognition task, and a simple yet effective method to address the scene bias for the task.

Mitigating the Human-Robot Domain Discrepancy in Visual Pre-training for Robotic Manipulation
Jiaming Zhou, Teli Ma, Kun-Yu Lin, Ronghe Qiu, Zifan Wang, Junwei Liang
arXiv, 2024
arXiv / project page

A new paradigm utilizing paired human-robot videos to adapt human-data pretrained models for robotic manipulation tasks.

Human-Centric Transformer for Domain Adaptive Action Recognition
Kun-Yu Lin, Jiaming Zhou, Wei-Shi Zheng
TPAMI, 2024
paper / arXiv

A human-centric video network to address the context bias in domain adaptive action recognition.

Diversifying Spatial-Temporal Perception for Video Domain Generalization
Kun-Yu Lin, Jia-Run Du, Yipeng Gao, Jiaming Zhou, Wei-Shi Zheng
NeurIPS, 2023
paper / arXiv / github

A diversity-aware video network to address the bias to domain-specific information in video domain generalization.

Event-Guided Procedure Planning from Instructional Videos with Text Supervision
An-Lan Wang#, Kun-Yu Lin#, Jia-Run Du, Jingke Meng, Wei-Shi Zheng
ICCV, 2023
paper / arXiv

A new event-guided paradigm to address the semantic gap between observed states and unobserved actions for procedure planning in instructional videos.

AsyFOD: An Asymmetric Adaptation Paradigm for Few-Shot Domain Adaptive Object Detection
Yipeng Gao#, Kun-Yu Lin#, Junkai Yan, Yaowei Wang, Wei-Shi Zheng
CVPR, 2023
paper / github

An asymmetric adaptation paradigm for few-shot domain adaptive object detection.

DilateFormer: Multi-Scale Dilated Transformer for Visual Recognition
Jiayu Jiao#, Yu-Ming Tang#, Kun-Yu Lin, Yipeng Gao, Jinhua Ma, Yaowei Wang, Wei-Shi Zheng
TMM, 2023
paper / arXiv / project page / github

A new vision transformer architecture for efficient and effective visual understanding.

Supervision Adaptation Balancing In-distribution Generalization and Out-of-distribution Detection
Zhilin Zhao, Longbing Cao, Kun-Yu Lin
TPAMI, 2023
paper / arxiv / github

A theorectical method to balancing in-distribution generalization and out-of-distribution detection.

Revealing the Distributional Vulnerability of Discriminators by Implicit Generators
Zhilin Zhao, Longbing Cao, Kun-Yu Lin
TPAMI, 2023
paper / arxiv / github

A theorectical method based on implicit generators to improve out-of-distribution detection.

Adversarial Partial Domain Adaptation by Cycle Inconsistency
Kun-Yu Lin, Jiaming Zhou, Yukun Qiu, Wei-Shi Zheng
ECCV, 2022
paper / github

A simple yet effective method based on cycle transformation to filter out outlier classes in partial domain adaptation.

Services

Reviewer of CVPR23, CVPR24
Reviewer of ICCV23
Reviewer of ECCV24
Reviewer of ICLR25
Reviewer of NeurIPS24
Reviewer of IJCAI24
Reviewer of TCSVT

This website borrows from Jon Barron.