Computer Vision

KIM TRAN

***************@*****.*** Linkedin LeetCode +1-479-***-****

EDUCATION

Master of Computer Science (Present), University of Arkansas, Fayetteville, United States. 2024–2026 GPA: 4.00/4.00

Bachelor of Computer Science, University of Science - VNUHCM, Ho Chi Minh City, Vietnam. 2019–2023 GPA: 8.95/10.00 (top 1% of Department), Thesis grade: 10.00/10.00 WORK EXPERIENCE

Research Assistant (Full Scholarship) June 2024 – Present EECS Department, University of Arkansas Fayetteville, US

• Researched and developed 3D multi-camera multi-object tracking solutions for autonomous driving and poultry science applications.

• Mastered camera calibration, SLAM, object detection and tracking algorithms. AI Research Resident July 2022 – May 2024

FPT Software Company, AI Center Ho Chi Minh, Vietnam

• Analyzed problems and provided solutions to computer vision challenges, including Multiple Object Tracking, Video Action Recognition, and Vision-Language models.

• Achieved four papers accepted at top conferences (see Publications section) in 2024. AI Engineer Intern April 2022 – July 2022

ELCA Company Ho Chi Minh, Vietnam

• Built a face recognition application on the company’s large image database (about 40,000 event images and 300 em- ployee profile photos). Processed real enterprise datasets with challenges such as illuminance, occlusion, viewpoint, and large-scale issues.

PERSONAL PROJECTS

Augmented Reality Indoor Navigation App

This project employs a deep-learning SLAM system to reconstruct a 3D mesh of the indoor scene. The mesh is precisely anchored to the physical environment, allowing for overlaying the AR path directly onto the user’s view. Deployed as an Android app, it empowers users to navigate complex indoor layouts such as airports, shopping centers, and office complexes with confidence and ease. (Demo) (GitHub) Augmented Reality Game

This project involves developing an augmented reality game. This game enables a virtual cannon to accurately fire virtual projectiles toward the player within the interactive 3D game scene. (Demo) SELECTED PUBLICATIONS

[NAACL’24] Zero-Shot Generic Multiple Object Tracking, by Kim Hoang Tran, Anh Duy Le Dinh, Tien Phat Nguyen, Ngan Hoang Le, et al. Long Paper, 2024 Annual Conference of the North American Chapter of the Associ- ation for Computational Linguistics, Mexico City, Mexico. (Proceedings) (Demo website)

[IJCNN’24] Unifying Global and Local Scene Entities Modelling for Precise Action Spotting, by Kim Hoang Tran, Phuc Do Vuong, Ngoc Ly Quoc, Ngan Hoang Le. Long Paper, 2024 International Joint Conference on Neural Networks, Yokohama, Japan. (Oral Presentation) (Proceedings) (GitHub)

[ECAI’24] TP-GMOT: Tracking Generic Multiple Object by Textual Prompt with Motion-Appearance Cost (MAC) SORT, by Duy Le Dinh Anh, Kim Hoang Tran, Ngan Hoang Le. Long Paper, 2024 European Conference on Ar- tificial Intelligence, Santiago de Compostela, Spain. (Oral Presentation) (Proceedings) (Demo website)

[ACCV’24] Enhanced Kalman with Adaptive Appearance Motion SORT for Grounded Generic Multiple Object Tracking, by Duy Le Dinh Anh, Kim Hoang Tran, Quang Thuc Nguyen, Ngan Hoang Le. Long Paper, 2024 Asian Conference on Computer Vision, Hanoi, Vietnam. (Oral Presentation) (Proceedings) SKILLS

Programming Languages: Python, C++, C#, SQL.

Frameworks: PyTorch, TensorFlow, NumPy, OpenCV, OpenGL, Unity, Spark, Hadoop.

Contact this candidate