Ming-Xuan LIU

Ming-Xuan LIU 

2nd-year PhD Student
Department of Information Engineering and Computer Science, University of Trento

Visiting Researcher
NAVER LABS Europe

🛠 I love Engineering (mechanical, electromechanical, and algorithmic) as much as I love Hotpot 🍲

I practice and support Slow Science, just because I need time to understand the problems that I study.

Publications | Patent | Projects | Google Scholar | Github | LinkedIn | Twitter | CV | CV of Failures | Services | Wolowitz

Email: mingxuan.liu@unitn.it (primary) | mingxuanliu.miu@gmail.com

About Me

I am a second-year PhD student in Deep Learning and Computer Vision at the University of Trento, under the joint supervision of Prof. Elisa RICCI and Prof. Zhun ZHONG. Currently, I am also a visiting researcher at NAVER LABS Europe exploring open-vocabulary object detection supervised by Gabriela CSURKA, Riccardo VOLPI, and Tyler L. HAYES.

My research is dedicated to improving machine Open-world Perception and Understanding through visual input. This endeavor sparked my keen interest in areas Multi-modal Learning (open-vocabulary object detection, vision-language), and Novel Knowledge Discovery and Reasoning (semi-/un-supervised learning, incremental learning, clustering, and multi-modal reasoning). Recently, my focus has particularly shifted towards exploring how the knowledge and reasoning capabilities of Large Language Models (LLMs) can help vision tasks in real-world scenarios.

Prior to commencing my doctoral studies, I cultivated a profound interest in Robotics, which led me to earn two Master's degrees in Autonomous Systems with distinction from KTH Royal Institute of Technology and the University of Trento. Before embarking on my academic journey, I gained valuable industry experience over three years as an Innovation Engineer at SIEMENS Smart Infrastructure Division. There, I played an active role in developing innovative automation solutions using Internet-of-Things (IoT) technologies.

In my free time, I enjoy bodybuilding, playing basketball, hiking, and cooking! Big fan of The Big Bang Theory and Eminem.

News

  • 10/2024: I'm excited to share that I'll be joining the Zhou Lab at UCLA as a visiting student from Jan-2025 for six months, where I'll have the privilege of being advised by Prof. Bolei Zhou . Looking forward to diving into some cool research (w. more more dimensions? : ) at Zhou Lab—and also hitting the Muscle Beach Venice Gym to lift some weights!

  • 10/2024: Our new work on automatically discovering grouping criteria and visual data semantic substructures are preprinted. Check it out here!

  • 09/2024: Happy to serve as a reviewer for ICLR 2025.

  • 07/2024: I successfully got $5,000 funding (in credits) from OpenAI to support my research!

  • 05/2024: One paper on incremental novel class discovery with large scale pre-trained model is accepted Oral paper at ICPR 2024..

  • 05/2024: I am invited to serve as a reviewer for NeurIPS 2024.

  • 05/2024: Filed my first US Patent: "A Method for Using Semantic Hierarchy Trees to Increase the Robustness of Open-vocabulary Object Detection Models"!

  • 02/2024: One paper on open-vocabulary object detection with semantic hierarchy (work done with NAVER LABS Europe) is accepted as Highlight paper (2.8% acceptance rate) at CVPR 2024! Thanks to the team!

  • 02/2024: I am invited to serve as a reviewer for ECCV 2024 and IJCV.

  • 01/2024: One paper on discovering fine-grained semantic concepts with LLMs is accepted to ICLR 2024! See you in Vienna this May!

Patent

A Method for Using Semantic Hierarchy Trees to Increase the Robustness of Open-vocabulary Object Detection Models
Mingxuan Liu, Tyler L. Hayes, Gabriela Csurka, Elisa Ricci, Riccardo Volpi
US Patent App. (status: filed; under processing)

Publications

Organizing Unstructured Image Collections using Natural Language
Mingxuan Liu, Zhun Zhong, Jun Li, Gianni Franchi, Subhankar Roy, Elisa Ricci
Preprint, 2024
[Paper] [Project Page] [Code (coming soon)]

 

SHiNe: Semantic Hierarchy Nexus for Open-vocabulary Object Detection
Mingxuan Liu, Tyler L. Hayes, Gabriela Csurka, Elisa Ricci, Riccardo Volpi
The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR, Highlight, 2.8% acceptance rate), 2024
[Paper] [Code]

 

Democratizing Fine-grained Visual Recognition with Large Language Models
Mingxuan Liu, Subhankar Roy, Wenjing Li, Zhun Zhong, Nicu Sebe, Elisa Ricci
International Conference on Learning Representations (ICLR), 2024
[Paper] [Project Page] [Code] [Poster]

 

Large-scale Pre-trained Models are Surprisingly Strong in Incremental Novel Class Discovery
Mingxuan Liu, Subhankar Roy, Zhun Zhong, Nicu Sebe, Elisa Ricci
International Conference on Pattern Recognition (ICPR, Oral), 2024
[Paper] [Code]

 

Class-incremental Novel Class Discovery
Subhankar Roy*, Mingxuan Liu*, Zhun Zhong, Nicu Sebe, Elisa Ricci
European Conference on Computer Vision (ECCV), 2022
[Paper] [Code] [Poster]

* = co-first author

 

Siemens Designo CC Building Management System Software and Practice
Huixia Zhao, Jiaxin Han, Kaixuan Zhang, Chao Wang, Lin Feng, Jianqiao Feng, Jian Li, Mingxuan Liu
China Electric Power Press (CEPP), 2023, ISBN: 9787519853341
[Book]

Tutorial book about building automation management software (Siemens Desigo CC). Work carried out at SIEMENS.

 

Siemens RWG Control Platform Advanced Course and Practice
Jiaxin Han, Huixia Zhao, Kaixuan Zhang, Mingxuan Liu
China Electric Power Press (CEPP), 2022, ISBN: 9787519859947
[Book]

Tutorial book about IoT-based building automation control platform, including Programmable Logic Controller (PLC), Internet-of-Things (IoT), and cloud-based Software-as-a-Service (SaaS). Work carried out at SIEMENS.

Projects

The V-SLAM Hurdler: A Faster V-SLAM System using Online Semantic Dynamic-and-Hardness-aware Approximation
Mingxuan Liu
[Master Thesis]
Digitala Vetenskapliga Arkivet (DiVA), 2022


A faster dynamic visual SLAM system with spatial-hardness-aware approximation. Worked carried at Ericsson, Lund, Sweden, as Thesis Intern.

ORB-SLAM3 Deployment on Underwater Autonomous Vehicle (UAV) SAM
Mingxuan Liu
[Code]
Work carried at KTH Royal Institute of Technology supervised by Prof. John Folkesson, 2021


A ORB-SLAM3 deployment on a small UAV robot via ROS.

Hot Steel Plate Tracking and Rotation Angle Detection
Mingxuan Liu, Polo Teta, Giulia Tucci
[Code]
Work carried at Technical University of Munich summer school for SMS Group, 2020


A YOLO-v5 based hot steel plate tracking and rotation angle detection algorithm for steel manufacturing factories.

Design, Kinematic and Dynamic Simulation of a Mini Cheetah Robotic Leg
Mingxuan Liu,
[Code]
Work carried at University of Trento supervised by Prof. Francesco Biral, 2020


A leg design for a lightweight quadruped robot (cheetah) with kinematic and dynamic simulation.

 

Virtual RGB-D Camera Unity Implementation
Mingxuan Liu,
[Code]
Work carried at University of Trento, 2020


An off-the-shelf virtual RGB-D camera implemented as Unity package for point cloud research.

Community Services

Acknowledgement

This webpage template was inspired by Prof. Jia-Bin Huang's personal webpage.