Hello, I'm Minh Luu (Nguyen Hoang)

I am a researcher in 3D computer vision, generative models, and computer graphics. I am currently a AI Resident at VinAI. I received my Bachelor of Computer Science from the Ho Chi Minh University of Technology in 2023.

CV Scholar Github Twitter

News

  • Feb 2025: Any3DIS has been accepted to CVPR 2025.
  • Nov 2024: Our new preprints ModeDreamer and Any3DIS are out.
  • Aug 2024: Our new preprint OE-3DIS is out.
  • Jul 2024: DiverseDream has been accepted to ECCV 2024.
  • Nov 2023: I earned my B.E. in Computer Science at HCMUT
  • Feb 2023: I have joined VinAI Research as a research resident.

Publications
* means equal contribution

Any3DIS: Class-Agnostic 3D Instance Segmentation by 2D Mask Tracking

Any3DIS: Class-Agnostic 3D Instance Segmentation by 2D Mask Tracking

Phuc Nguyen, Minh Luu, Anh Tran, Cuong Pham Nguyen, Khoi Nguyen
CVPR, 2025

A novel class-agnostic approach for 3D instance segmentation that leverages 2D mask tracking to segment 3D objects in point cloud scenes.

Project Page arXiv
DiverseDream: Diverse Text-to-3D Synthesis with Augmented Text Embedding

DiverseDream: Diverse Text-to-3D Synthesis with Augmented Text Embedding

Minh Luu*, Uy Tran*, Phong Nguyen, Khoi Nguyen, Binh-Son Hua
ECCV, 2024

Tackling mode-collapses problem in existing text-to-3D generation by using personalized text prompts to diversify the joint generation.

Project Page arXiv
ModeDreamer: Mode Guiding Score Distillation for Text-to-3D Generation using Reference Image Prompts

ModeDreamer: Mode Guiding Score Distillation for Text-to-3D Generation using Reference Image Prompts

Minh Luu*, Uy Tran*, Phong Nguyen, Khoi Nguyen, Binh-Son Hua
Arxiv Preprint, 2024

Introducing a novel Image Prompt Score Distillation (ISD) loss to address over-smoothing and low-quality outputs in text-to-3D generation caused by mode-seeking behavior in SDS-based method.

Project Page arXiv
OE-3DIS: Open-Ended 3D Point Cloud Instance Segmentation

OE-3DIS: Open-Ended 3D Point Cloud Instance Segmentation

Minh Luu*, Phuc Nguyen*, Anh Tran, Cuong Pham Nguyen, Khoi Nguyen
Arxiv Preprint, 2024

Introducing the Vocablulary-Free 3D point cloud instance segmentation with different solid baselines and a novel pointwise method using multimodal LLM.

arXiv