Tuyen Tran

Applied Artificial Intelligence Institute (A2I2), Deakin University.

prof_pic.jpg
Me at IJCAI 2024 held in Jeju Island, South Korea.

Hello! I’m Tuyen Tran and welcome to my homepage. I am a Ph.D student in Computer Vision at Applied Artificial Intelligence Institute (A2I2), Deakin University, based in Australia. My research interest is about Video Understanding, especially on Action Recognition, Referring Video Object Segmentation, and Video Event Understanding.

I use this homepage to share my research activities. I also write about topic that find interesting, which would be about Computer Vision, Machine Learning, and sometimes about my personal life.

News

Jul 10, 2025 Our paper Planner-Refiner: Dynamic Space-Time Refinement for Vision-Language Alignment in Videos has been accepted for presentation at the ECAI 2025 (Acceptance rate: 23%, 626/2,667 submissions).
Jul 10, 2025 Our paper Towards Agentic AI for Multimodal-Guided Video Object Segmentation has been accepted for presentation at the MMFM4: The 4th Workshop on What is Next in Multimodal Foundation Models? at ICCV 2025 Workshop.
Aug 15, 2024 I achieved second place in the ECCV 2024 6th LSVOS Challenge.
Aug 09, 2024 Our paper Promptable Iterative Visual Refinement for Video Instance Segmentation has been accepted for presentation at the Instance-Level Recognition Workshop at ECCV 2024.
Jul 25, 2024 Our paper Unified Framework with Consistency across Modalities for Human Activity Recognition has been accepted for presentation at the BMVC 2024 (Acceptance rate: 25.8%, 246/1,020 submissions).
Jun 02, 2023 I achieved second place in the CVPR 2023 OmniLabel Challenge, with an award of $3,300 USD.

Selected publications

  1. ECAI
    ecai.png
    Planner-Refiner: Dynamic Space-Time Refinement for Vision-Language Alignment in Videos
    Tuyen Tran, Thao Minh Le, Le Hung, and 1 more author
    In European Conference on Artificial Intelligence (ECAI), 2025
    arXiv
  2. BMVC
    bmvc.png
    Unified Framework with Consistency across Modalities for Human Activity Recognition
    Tuyen Tran, Thao Minh Le, Hung Tran, and 1 more author
    In British Machine Vision Conference (BMVC), 2024
    arXiv Video Poster
  3. LSVOS Challenge
    eccv.png
    The 2nd Solution for LSVOS Challenge RVOS Track: Spatial-temporal Refinement for Consistent Semantic Segmentation
    Tuyen Tran
    In ECCV Workshops, 2024
    arXiv