Wei Lin

I am a research associate at the Institute for Machine Learning headed by Prof. Sepp Hochreiter (Father of LSTM) at the Johannes Kepler University (JKU) Linz.

I did my PhD at the Institute of Computer Graphics and Vision, Graz University of Technology (TU Graz ICG) in Austria, supervised by Prof. Horst Bischof (Professor and the Rector of TU Graz) and Prof. Hilde Kuehne (Tuebingen AI Center, University of Tuebingen, MIT-IBM Watson AI Lab)

I also work in close collaboration with Leonid Karlinsky and Rogerio Feris (Principal scientists and research manager) from the MIT-IBM Watson AI Lab.

Previously, I received my Master's degree in Electrical and Computer Engineering at the Technical University of Munich in Germany.

I like traveling 🥾 and learning languages 🌍 I speak English (C1), German (B2), Chinese (mother tongue) and some French.

Email  /  CV  /  Google Scholar  /  Github  /  LinkedIn  /  Twitter

profile photo
News Research

I am interested in computer vision and machine learning in general. My research is mainly about vision-language models, multimodal learning, domain adaptation and video understanding.

Comparison Visual Instruction Tuning
Wei Lin, Muhammad Jehanzeb Mirza, Sivan Doveh,
Rogerio Feris, Raja Giryes, Sepp Hochreiter,
Leonid Karlinsky
In collaboration with the MIT-IBM Watson AI Lab
Arxiv, 2024
arxiv / 🤗 Dataset / code / video

an approach for collection of visual instructions that improves Commonality and difference spoting capabilities for Large Multimodal Modes

Conme: Rethinking Evaluation of Compositional Reasoning for Modern VLMs
*Irene Huang, *Wei Lin, *Muhammad Jehanzeb Mirza, Jacob Hansen, Sivan Doveh,
Victor Ion Butoi, Roei Herzig, Assaf Arbelle, Hilde Kuehne, Trevor Darrell, Chuang Gan, Aude Oliva, Rogerio Feris, Leonid Karlinsky (*equal contribution)
In collaboration with the MIT-IBM Watson AI Lab
NeurIPS, 2024 Datasets & Benchmarks Track
arxiv / 🤗 Dataset / code / video

Meta-Prompting for Automating Zero-shot Visual Recognition with LLMs
Muhammad Jehanzeb Mirza, Leonid Karlinsky, Wei Lin, Sivan Doveh,
Jakub Micorek, Mateusz Kozinski, Hilde Kuehne, Horst Possegger
In collaboration with the MIT-IBM Watson AI Lab
ECCV, 2024
arxiv / code / video

Vision-Language Guidance for LiDAR-based Unsupervised 3D Object Detection
Christian Fruhwirth-Reisinger, Wei Lin, Dusan Malic,
Horst Bischof, Horst Possegger
BMVC, 2024 Oral Presentation
arxiv / code / video

Towards Multimodal In-Context Learning for Vision & Language Models
Sivan Doveh, Shaked Perek, Muhammad Jehanzeb Mirza,
Wei Lin, Amit Alfassy, Assaf Arbelle,
Shimon Ullman, Leonid Karlinsky
ECCV 2024 Workshop on Multimodal Agents
arxiv / code / video

Overlooked Aspects in the Evaluation of Out-Of-Distribution Detection Methods
*Bernhard Lehner, *Christian Huber, Bernhard Moser,
Claus Hofmann, Wei Lin, Sepp Hochreiter (*equal contribution)
Arxiv, 2024
arxiv / code / video

LaFTer: Label-Free Tuning of Zero-shot Classifier using Language and Unlabeled Image Collections
Muhammad Jehanzeb Mirza, Leonid Karlinsky, Wei Lin,
Mateusz Kozinski, Horst Possegger, Rogerio Feris, Horst Bischof
NeurIPS, 2023
arxiv / code / video

MAtch, eXpand and Improve: Unsupervised Finetuning for Zero-Shot Action Recognition with Language Knowledge
Wei Lin, Leonid Karlinsky, Nina Shvetsova,
Horst Possegger, Mateusz Kozinski, Rameswar Panda,
Rogerio Feris, Hilde Kuehne, Horst Bischof
In collaboration with the MIT-IBM Watson AI Lab
ICCV, 2023
arxiv / code / video

Unsupervised finetuning of Vision-Language models for zero-shot and few-shot action recognition, with GPT3 text expansion and video frame captioning.

TAP: Targeted Prompting for Task Adaptive Generation of Textual Training Instances for Visual Classification
Muhammad Jehanzeb Mirza, Leonid Karlinsky, Wei Lin,
Horst Possegger, Rogerio Feris, Horst Bischof
Arxiv, 2023
arxiv / code / video

MATE: Masked Autoencoders are Online 3D Test-Time Learners
*Muhammad Jehanzeb Mirza, *Inkyu Shin, *Wei Lin,
Andreas Schriebl, Kunyang Sun, Jaesung Choe,
Horst Possegger, Mateusz Kozinski, In So Kweon,
Kun-Jin Yoon, Horst Bischof (*equal contribution)
ICCV, 2023
arxiv / code / video

Video Test-Time Adaptation for Action Recognition
*Wei Lin, *Muhammad Jehanzeb Mirza, Mateusz Kozinski,
Horst Possegger, Hilde Kuehne, Horst Bischof (*equal contribution)
CVPR, 2023
arxiv / 🤗 Dataset / code / video

Test-time adaptation of video action recognition against common distribution shifts.

ActMAD: Activation Matching to Align Distributions for Test-Time-Training
Muhammad Jehanzeb Mirza, Pol Jané Soneira, Wei Lin,
Mateusz Kozinski, Horst Possegger, Horst Bischof
CVPR, 2023
arxiv / code / video

Unsupervised Class-aware 3D Object Detection in LiDAR Point Clouds
Christian Fruhwirth-Reisinger, Wei Lin, Dusan Malic,
David Schinagl, Georg Krispel, Horst Possegger, Horst Bischof
Arxiv, 2023
arxiv / code / video

CycDA: Unsupervised Cycle Domain Adaptation to Learn from Image to Video
Wei Lin, Anna Kukleva, Kunyang Sun,
Horst Possegger, Hilde Kuehne, Horst Bischof
ECCV, 2022
paper / arxiv / code / video

Unsupervised image-to-video domain adaptation.

Extended Abstract CycDA: Unsupervised Cycle Domain Adaptation to Learn from Image to Video
Wei Lin, Anna Kukleva, Kunyang Sun,
Horst Possegger, Hilde Kuehne, Horst Bischof
ECCV Workshop of Out Of Distribution Generalization in Computer Vision, 2022
paper / code / video

AIR-DA: Adversarial Image Reconstruction for Unsupervised Domain Adaptive Object Detection
Kunyang Sun, Wei Lin, Haoqin Shi,
Zhengming Zhang, Yongming Huang, Horst Bischof
IEEE Robotics and Automation Letters (RA-L) 2023
paper / arxiv / code / video

TAEC: Unsupervised Action Segmentation with Temporal-Aware Embedding and Clustering
Wei Lin, Anna Kukleva, Horst Possegger,
Hilde Kuehne, Horst Bischof
Computer Vision Winter Workshop, 2023
arxiv / code / video

Sit Back and Relax: Learning to Drive Incrementally in All Weather Conditions
Stefan Leitner, Muhammad Jehanzeb Mirza, Wei Lin,
Jakub Micorek, Marc Masana, Mateusz Kozinski,
Horst Possegger, Horst Bischof
Intelligent Vehicle Conference, 2023
arxiv / code / video

Academic Service
  • Conference Reviewer: ECCV 2022, ISMAR 2023, CVPR 2023, NeurIPS 2023, WACV 2024, CVPR 2024, ECCV 2024, NeurIPS 2024, NeurIPS 2024 Dataset and Benchmark Track
  • Journal Reviewer: TPAMI 2023, TNNLS 2023, IEEE Trans. Multimedia 2023, Pattern Recognition Letters 2024
Teaching
  • Deep Learning and Neural Networks I - Exercise
  • Machine Learning: Supervised Techniques - Exercise
  • Deep Learning and Neural Networks II - Exercise
  • Machine Learning: Unsupervised Techniques - Exercise
Activity

template from Jon Barrion