Thanh-Dat Truong
Postdoctoral Fellow
Also affiliated: Vietnam National University Ho Chi Minh City (2018–2019); Ho Chi Minh City University of Science (2018–2019)
Research Areas
Links
Biography and Research Information
OverviewAI-generated summary
Thanh-Dat Truong's research focuses on the application of machine learning and deep learning techniques to various analytical challenges. His work includes developing advanced models for action recognition, domain adaptation in semantic scene segmentation, and fairness in semantic scene understanding. Truong has also investigated the use of graph convolutional neural networks for movement analysis in neurological and musculoskeletal disorders, and lightweight deep convolutional networks for tiny object recognition.
His publications also address vehicle re-identification and abnormality detection for traffic video analysis, as well as traffic flow analysis and velocity estimation. Furthermore, Truong has explored smart lifelog retrieval systems utilizing habit-based concepts and moment visualization. He has a publication record of 64 papers, with 433 citations, and an h-index of 11. Truong collaborates with researchers from the University of Arkansas at Fayetteville, including Ky Luu, Ashley P. G. Dowling, Xuan-Bac Nguyen, and Jackson Cothren.
Metrics
- h-index: 11
- Publications: 64
- Citations: 433
Positions
-
Postdoctoral Fellow 2025–presentUniversity of Arkansas at Fayetteville Electrical Engineering and Computer Science ORCID
-
Postdoctoral Fellow publications 2022–2026Arkansas State University ORCID
Selected Publications
-
$ϕ$-DPO: Fairness Direct Preference Optimization Approach to Continual Learning in Large Multimodal Models (2026)arXiv (Cornell University) OpenAlex
-
MANGO: Multimodal Attention-based Normalizing Flow Approach to Fusion Learning (2025)
-
DEGA: Dynamic Entropy Guided Adaptation (2025)
-
Directed-Tokens: A Robust Multi-Modality Alignment Approach to Large Language-Vision Models (2025)
-
Insect-Foundation: A Foundation Model and Large Multimodal Dataset for Vision-Language Insect Understanding (2025)
-
Cross-view action recognition understanding from exocentric to egocentric perspective (2024)
-
CONDA: Continual Unsupervised Domain Adaptation Learning in Visual Perception for Self-Driving Cars (2024)
-
Insect-Foundation: A Foundation Model and Large-Scale 1M Dataset for Visual Insect Understanding (2024)
-
LIAAD: Lightweight attentive angular distillation for large-scale age-invariant face recognition (2023)
-
OTAdapt: Optimal Transport-based Approach For Unsupervised Domain Adaptation (2022)
-
Self-Supervised Domain Adaptation in Crowd Counting (2022)
-
EQAdap: Equipollent Domain Adaptation Approach to Image Deblurring (2022)
-
BiMaL: Bijective Maximum Likelihood Approach to Domain Adaptation in Semantic Scene Segmentation (2021)
-
Movement Analysis for Neurological and Musculoskeletal Disorders Using Graph Convolutional Neural Network (2021)
-
Fast Flow Reconstruction via Robust Invertible n × n Convolution (2021)
Collaboration Network
Top Collaborators
- Movement Analysis for Neurological and Musculoskeletal Disorders Using Graph Convolutional Neural Network
- Insect-Foundation: A Foundation Model and Large-Scale 1M Dataset for Visual Insect Understanding
- LIAAD: Lightweight attentive angular distillation for large-scale age-invariant face recognition
- EQAdap: Equipollent Domain Adaptation Approach to Image Deblurring
- Self-Supervised Domain Adaptation in Crowd Counting
Showing 5 of 15 shared publications
- LIAAD: Lightweight attentive angular distillation for large-scale age-invariant face recognition
- Vec2Face: Unveil Human Faces From Their Blackbox Features in Face Recognition
- Fast Flow Reconstruction via Robust Invertible n × n Convolution
- DyGLIP: A Dynamic Graph Model with Link Prediction for Accurate Multi-Camera Multiple Object Tracking
- Generative Flow via Invertible nxn Convolution.
Showing 5 of 6 shared publications
- LIAAD: Lightweight attentive angular distillation for large-scale age-invariant face recognition
- Self-Supervised Domain Adaptation in Crowd Counting
- OTAdapt: Optimal Transport-based Approach For Unsupervised Domain Adaptation
- Fast Flow Reconstruction via Robust Invertible n × n Convolution
- Generative Flow via Invertible nxn Convolution.
Showing 5 of 6 shared publications
- Vehicle Re-identification with Learned Representation and Spatial Verification and Abnormality Detection with Multi-Adaptive Vehicle Detectors for Traffic Video Analysis
- Fast Flow Reconstruction via Robust Invertible n × n Convolution
- DyGLIP: A Dynamic Graph Model with Link Prediction for Accurate Multi-Camera Multiple Object Tracking
- Generative Flow via Invertible nxn Convolution.
- LIAAD: Lightweight attentive angular distillation for large-scale age-invariant face recognition
- Vec2Face: Unveil Human Faces From Their Blackbox Features in Face Recognition
- DyGLIP: A Dynamic Graph Model with Link Prediction for Accurate Multi-Camera Multiple Object Tracking
- Insect-Foundation: A Foundation Model and Large-Scale 1M Dataset for Visual Insect Understanding
- Insect-Foundation: A Foundation Model and Large Multimodal Dataset for Vision-Language Insect Understanding
- OTAdapt: Optimal Transport-based Approach For Unsupervised Domain Adaptation
- Self-Supervised Domain Adaptation in Crowd Counting
- DyGLIP: A Dynamic Graph Model with Link Prediction for Accurate Multi-Camera Multiple Object Tracking
- Movement Analysis for Neurological and Musculoskeletal Disorders Using Graph Convolutional Neural Network
- EQAdap: Equipollent Domain Adaptation Approach to Image Deblurring
- EQAdap: Equipollent Domain Adaptation Approach to Image Deblurring
- BiMaL: Bijective Maximum Likelihood Approach to Domain Adaptation in Semantic Scene Segmentation
- EQAdap: Equipollent Domain Adaptation Approach to Image Deblurring
- CONDA: Continual Unsupervised Domain Adaptation Learning in Visual Perception for Self-Driving Cars
- Insect-Foundation: A Foundation Model and Large-Scale 1M Dataset for Visual Insect Understanding
- OTAdapt: Optimal Transport-based Approach For Unsupervised Domain Adaptation
- Insect-Foundation: A Foundation Model and Large-Scale 1M Dataset for Visual Insect Understanding
- Insect-Foundation: A Foundation Model and Large Multimodal Dataset for Vision-Language Insect Understanding
- MANGO: Multimodal Attention-based Normalizing Flow Approach to Fusion Learning
- $ϕ$-DPO: Fairness Direct Preference Optimization Approach to Continual Learning in Large Multimodal Models
- Vehicle Re-identification with Learned Representation and Spatial Verification and Abnormality Detection with Multi-Adaptive Vehicle Detectors for Traffic Video Analysis
- Vehicle Re-identification with Learned Representation and Spatial Verification and Abnormality Detection with Multi-Adaptive Vehicle Detectors for Traffic Video Analysis
Similar Researchers
Based on overlapping research topics