Umar Farooq

Umar Farooq

AI Researcher

Hanyang University, Seoul, South Korea

Mon-Fri (10:00 ~ 19:00) KST

About Me

I am an AI Researcher in the Department of Computer Science at Hanyang University Seoul campus, South Korea, where I focus on cutting-edge AI projects, including image classification, object detection, and segmentation. I received my MS degree in Artificial Intelligence from the Korea National University of Transportation in Chungju, where I studied under the supervision of Prof. Jeonghwan Gwak. Before joining Hanyang University, I worked as an AI researcher at the Applied Machine Intelligence Lab and a small Korean AI company Decasight, engaging in various deep learning-based projects such as medical image classification, segmentation, surgical instrument tip tracking using AR, VR, XR and other Computer-Aided Diagnosis (CAD) systems.

My research interests are broadly in the area of machine/deep learning, computer vision and AI. Recent several years, my research interest was mainly focuse on knowledge distillation, semi-supervised learning, self-supervised learning, active learning, contrastive learning, etc. In addition, I have also conducted extensive researches on CADx and CADe systems.

If you have any query, please drop me an email: My Email

Contact: umarfarooq at hanyang dot ac dot kr
Office Location: Research and Development Building, Hanyang University, Seoul, Republic of Korea.

Interests

  • Augmented Reaity
  • Surgical Instrument Navigation using AR
  • Semi-Supervised Learning
  • Self-supervised Learning
  • Active learning
  • Knowledge Distilation
  • CAD systems

Education

  • Ph.D in AI, 2023~

    Department of Computer Science, Hanyang University, Seoul, South Korea

  • M.S in Artificial Intelligence, 2023

    Korea National University of Transportatoin, Chungju, South Korea

News

  • July. 2024: First author paper title as "GDSSA-Net: A Gradually Deeply Supervised Self-ensemble Attention Network for IoMT-Integrated Thyroid Nodule Segmentation" has been submitted to Internet of Things journal (Elsevier). (SCI, JCR 2021 Edition = IF 6.0
  • Paper title as "GDSSA Net: A Gradually Deep Supervised, Self-Ensemble Attention Network for IoMT-Integrate Thyroid Nodule Segmentation" is Under Submission, Internet of Things journal (Elsevier).
    • .

Research Colaboration Team

  • Mr. Muhammad Mutti Ur Rehman Undergrad student at National University of Science and Technology, Islamabad, Pakistan
  • Mr. Haris Ghafoor Department of Computational Science and Engineering, Yonsei University, Seoul, South Korea
  • Azka Rehman MS student at Seoul National University, Seoul, South Korea
  • Dr. Muhammad Usman Postdoctoral Researcher at Stanford University, Stanford, USA.
  • Dr. Abdur Rehman: Research Scientist at Donggok University, Seoul, South Korea.
  • Mr. Inam ul Haq: Senior web Design and Developer at IZZ web solution, Islamabad, Pakistan. .
  • Mr. Saeed Ahmed Khan: Master Student and Research assistant at Korea Nationa University of Transportation, South Korea.
  • Mr. Huxufeng: Ph.D Student and AI Research Scientist at Korea National University of Transportation, South Korea.
  • Professor: Assistant Professor in the Department of Computer Science and Department of Artificial Intelligence at Hanyang University, Seoul, South Korea.
  • Professor Jeonghwan Gwak: Assistant Professor at the Department of Software, Korea National University of Transportation, Chungju-si, South Korea.
  • Dr Zahid Ullah: Postdoctoral Researcher at the Department of Software, Korea National University of Transportation, Chungju-si, South Korea.

Work Experience

 
 
 
 
 

AI Researcher

Hanyang University, Seoul, South Korea

Feb 2023 – Present Seoul, South Korea
Cutting-edge AI Research on self-supervised learning for images.
Research by following updates via channels such as Twitter, Reddit, Arxiv, and conferences such as NeurIPS, ICLR, ICML, CVPR, ICCV and ECCV.
 
 
 
 
 

Student Researcher

Applied Machine Intelligence Lab

I proposed a new CAD system that was published in Computerized Medical Imaging and Graphics journal supported by Brain Korea (BK21).
I worked on medical image diagnosis methods with limited data and resources with Prof. Jeonghwan Gwak.
 
 
 
 
 

AI Research Engineer

Healthhub, Seoul

Feb 2020 – Jul 2020 Seoul, South Korea
Role: Worked on a range of AI‑based research projects, such as liver tumor segmentation in CT scans, and successfully integrated the developed tools as part of the R&D team.
 
 
 
 
 

Sr. Customer Support Engineer

Team Lead of Customer Support Engineering at Wavetec Private Ltd.

April 2019 – Feb 2021 Jhelum, Pakistan
Provide technical support and assistance to customers, resolve complex issues, and ensure customer satisfaction.
Work closely with the sales and product development teams to gather customer feedback and improve products.
Training of new employees in customer support procedures and providing mentorship to junior team members.
 
 
 
 
 

Web Design and Developer

HA Technologies (Pvt) Ltd.

Sep 2018 – Mar 2019 Islamabad, Pakistan
Create a visual layout and user interface of a website, using design tools such as Adobe Photoshop and Illustrator. And use HTML and CSS to create web‑ready graphics and animations.
Design and develop functional components of a website such as a backend database and server‑side scripting
Using JavaScript, PHP, and Python create web applications and implement interactivity and functionality on the website

Publications

Please see my Google Scholar for the complete publication list.

Residual attention based uncertainty‑guided mean teacher model for semi‑supervised breast masses segmentation in 2D ultrasonography

[Computerized Medical Imaging and Graphics] We presented a deep learning-based approach for semi-supervised segmentation of breast masses in 2D ultrasound images.
Umar Farooq, Zahid Ulla, Jeonghwan Gwa

A hybrid image enhancement based brain MRI images classification technique

[Medical Hypotheses ]We have proposed a novel approach for classifying brain MRI images using a combination of image enhancement and machine learning techniques.
Umar Farooq, Zahid Ulla, Jeonghwan Gwa

Dual channel adversarial autoencoder with multitask learning for KL grade classification in knee radiographs

[Computers in Biology and Medicine (Under review) We have proposed a deep learning-based approach for classifying knee radiographs into different Kellgren-Lawrence (KL) grades. .
Umar Farooq, Zahid Ulla, Jeonghwan Gwa

Honors & Awards

  • Brain Korea 21(BK21) Scholarship 2020‑2021 Brain Korea 21 financial support
  • Full Tuition Fee Waiver Department of Artificial Intelligence, Korea National University of Transportation, Chungju
  • Prof. Stipend, Applied Machine Intelligence Lab, Korea National University of Transportation, Chungju
  • Lab Support Grants: Funding used to buy necessary resources such as equipment, supplies, and training for research and experimentation in the laboratory.
  • KNUT Scholarship: Excellent Performance Scholarship.
  • PEEF Scholarship Award: Highly Competitive Student Scholarship PEEF (Punjab Educational Endowment Fund) Scholarship [2016‑2018]
  • Prime Minister Pakistan Program, Islamabad, Punjab, Pakistan
  • FUUAST: Excellent Performance Award from Federal Urdu University of Arts, Science and Technology, Islamabad [2014‑2015]

Contact