Duc Hai Pham

I am a Research Intern at Qualcomm AI Research in Vietnam.

Until recently, I worked at VinAI Research, under the supervision of Dr. Rang Nguyen, I tried to make 3D scene understanding more affordable (by using less supervision) and more practical (by being more robust to real-world data). I have had the pleasure to collaborate with Prof. Son Hua, Dr. Phong Nguyen and Dr. Khoi Nguyen .

I obtained a Bachelors in Computer Engineering at Vietnam National University - University of Technology

Email  |  CV  |  GitHub  |  Google Scholar  |  LinkedIn  |  Twitter

headshot
Qualcomm VinAI HCMUT
News
  • 04/2025 Career Update: I start working as a research intern at Qualcomm AI Research!
  • 02/2025 Our paper SharpDepth is accepted to CVPR 2025! I'll be at Nashville to present (let's meet up).
  • 12/2024 Our paper VFG-SSC is accepted to AAAI 2025!
Research
WildGaussians GIF SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation
Duc-Hai Pham*, Tung Do*, Phong Nguyen, Binh-Son Hua, Khoi Nguyen, Rang Nguyen
Conference on Computer Vision and Pattern Recognition (CVPR), 2025
paper | project page |

A diffusion depth model for refining metric depth estimators to have more sharp depth boundaries, improving details in thin structures and point cloud quality.

WildGaussians GIF Semi-supervised 3D Semantic Scene Completion with 2D Vision Foundation Model Guidance
Duc-Hai Pham, Tuan Ho, Duc Dung Nguyen, Anh Pham, Phong Nguyen, Khoi Nguyen, Rang Nguyen
The 39th Annual AAAI Conference on Artificial Intelligence (AAAI), 2025
paper | project page |

A semi-supervised framework for 3D semantic occupancy prediction that reduces reliance on costly labeled data by leveraging 2D foundation models, achieving up to 85% of full-supervision performance with only 10% labeled data.

Selected Projects
WildGaussians GIF

3D Semantic Occupancy from Surrounding View Images
Tuan Ngo, Duc-Hai Pham, Tuan Ho, Khoi Nguyen, Rang Nguyen
CVPR 3D Occupancy Prediction Challenge, 2023
leaderboard |

Predict the 3D occupancy (semantics and geometry) of the scene from multi-view images. Ranked 16/200 teams, highest ranking Southest Asia Team.

Academic Services
  • Conference Reviewer: ACCV (2024), CVPR (2025)

template adapted from this awesome website


▶ haiphamcse.github.io's clustrmaps 🌎