Research
I have worked on research problems as follows:
Computer Vision, Generative AI, Extended Reality, Human-Computer Interaction
Applied Deep Learning to Computer Vision and Computer Graphics
Applied Computer Vision to Intelligent Transportation Systems, Multimedia Security, and Medical Imaging
I currently lead the Multimedia Forensics and Security (MFS) research group, focusing on issues of multimedia security and forensics. The team aims to protect human from malicious media synthesized by artificial intelligence (AI). We also study good and useful applications of AI-based synthetic media.
*** This page is no longer maintained. Please access the publication site. ***
Computer Vision
Real-time Virtual Try-On (ISMAR 2023)
Virtual Interior Design (ISMAR 2023)
Open Forensics Dataset (ICCV 2021)
Fashion-Guided Adversarial Attack (CVPRW 2021)
Camouflaged Instance Segmentation
CAMO++ Dataset (TIP 2022)
Live Demo (AAAI 2021)
Camouflaged Object Segmentation
CAMO Dataset (CVIU 2019)
MirrorNet (IEEE Access 2021)
ANet (CVIU 2019)
Semi-Supervised Video Object Segmentation
4th place, CVPR DAVIS Challenges 2020
3rd place, CVPR DAVIS Challenges 2019
6th place, CVPR DAVIS Challenges 2018
3rd place, CVPR DAVIS Challenges 2017
Wonderland Augmented Dataset (CVPRW 2018)
Interactive Video Object Mask Annotation (AAAI 2021)
Video Object Interactive Self-Annotation (WACV 2020)
Accident Detection In Dashboard Cameras
ADV Dataset (IV 2020)
Attention R-CNN (IV 2020)
Anomaly Detection In Surveillance Cameras
5th place, CVPR AI City Challenge 2020
8th place, CVPR AI City Challenge 2019
Vehicle Re-Identification
Rank 26/59, CVPR AI City Challenge 2020
Rank 25/84, CVPR AI City Challenge 2019
Video Semantic Salient Instance Segmentation
SESIV Dataset (WACV 2019)
Video Salient Object Segmentation
[3D Saliency] (BMVC 2017)
Text-to-Image Generation (ACM MM 2020)
Video Mask R-CNN
Artistic Style Transfer (WACV 2018)
Augmented Reality and Human-Computer Interaction (ICARCV 2014)
Augmented Reality and Visual Search (PACIS 2012)
In-Place Virtual Jogging (RIVF 2012)
My first research paper. I jointly worked with Quoc-Minh Bui. We propose a method to analyse differences in motion from consecutive frames to estimate person’s intensity level of activities when jogging, and then renders virtual scene with appropriate speed. Our proposed method can also be applied to develop various exciting interactive games to stimulate excitement for users when doing in-door fitness exercises or to create virtual environment for museums where visitors just stay in place to explore scenes in museums.
Security and Cryptography
Email Secure for PocketPC (Junior Research Award 2010)
My very first research project. I jointly worked with Nam Le, under supervision of Prof. Minh-Triet Tran. We studied two symmetric and asymmetric encryption methods, including Rijndael and Elliptic Curve Cryptography (ECC), to develop an application for encrypting emails in PocketPC.