Computer Vision& Retrieval.
A research group at the University of Information Technology, VNU-HCM. We build models that see, read, and retrieve — from multimodal interactive video retrieval to large-scale visual search.
DeadlineECIR · Oct 5, 2026Models that see, read, and retrieve.
CORE Lab is a computer-vision and information-retrieval research group founded in 2016 at the University of Information Technology, VNU-HCM. Our work spans multimodal interactive video retrieval, vision-language representation learning, scene-text understanding and content-based retrieval at scale.
Towards Scalable and Context-Aware Multimodal Interactive Video Retrieval
Bao Tran, Khiem Le, Thanh Duc Ngo
When Helpful Text Hurts: Option-Redirecting Bias in Vision–Language Models
Tam Le Thi Thanh, Hoang Tran Van, Thanh Duc Ngo
From Dialogue to Evidence: Retrieval-State-Conditioned Interaction for Text-Based Person Retrieval
Bao Tran, Thanh Duc Ngo
Research areas
Visual recognition, detection and representation learning.
Real-time, large-scale search over image and video archives.
Vision–language fusion across text, speech and audio.
Reading text in images, signage and broadcast video.
Tracking many objects across frames and camera views.
Interactive, dialogue-driven search for the right image.
Members
Latest
- Sep 1, 2025CORE Lab website goes live
We launched the CORE Lab website to share our research on computer vision and retrieval.
- Feb 1, 2025First place at the Video Browser Showdown (VBS) 2025
Our multimodal interactive video retrieval framework placed first at VBS 2025, ahead of all established systems.
- Dec 1, 2024Three consecutive wins at the Ho Chi Minh AI Challenge
CORE Lab secured three straight first-place finishes at the Ho Chi Minh AI Challenge (2022–2024) on multilingual broadcast video.
Interested in working with us?
We welcome students and collaborators in computer vision and retrieval. Reach out to the team.