Van Nguyen NGUYEN

I am a last-year PhD student at IMAGINE team, École des Ponts ParisTech, advised by Prof. Vincent Lepetit. I also work closely with Dr. Mathieu Salzmann and Dr. Thibault Groueix.

I graduated from the Master Mathématiques, Vision and Apprentissage (MVA) from École Normale Supérieure Paris-Saclay and Engineering Program in Applied Mathematics at INSA Toulouse. I also had chance to spend time with great teams at Meta Reality Labs, EPFL, SIEMENS.

Email  /  CV  /  Google Scholar  /  Linkedin  /  Twitter  /  Github

profile photo
News
Research

I am interested in 6D object pose estimation. My current research focuses on scalable 3D object detection and 6D pose estimation to handle arbitrary, previously-unseen objects.

GigaPose: Fast and Robust Novel Object Pose Estimation via One Correspondence
A "hybrid" template-patch correspondence approach that is fast, robust, and more accurate to estimate 6D pose of novel objects in RGB images. GigaPose predicts 6D object pose from a single 2D-to-2D correspondence.
Van Nguyen Nguyen, Thibault Groueix, Mathieu Salzmann, Vincent Lepetit
CVPR, 2024
project page arXiv github

Star Fork

NOPE: Novel Object Pose Estimation from a Single Image
A method that can estimate relative pose of unseen objects given only a single reference image. Our method also predicts 3D pose distribution which can be used to address pose ambiguities due to symmetries.
Van Nguyen Nguyen, Thibault Groueix, Georgy Ponimatkin, Yinlin Hu, Renaud Marlet, Mathieu Salzmann, Vincent Lepetit
CVPR, 2024
project page arXiv github

Star Fork

CNOS: A Strong Baseline for CAD-based Novel Object Segmentation
A method that can segment novel objects for a given RGB image from only their CAD models. Based on Segmenting Anything, DINOv2, CNOS is a strong baseline for Task 5 and 6 in the BOP challenge 2023.
Van Nguyen Nguyen, Thibault Groueix, Georgy Ponimatkin, Vincent Lepetit, Tomáš Hodaň
ICCV 2023 R6D Workshop, 2023
project page arXiv github

Star Fork

PIZZA: A Powerful Image-only Zero-Shot Zero-CAD Approach to 6DoF Tracking
A method for tracking the 6D motion of objects in RGB video sequences when neither training images nor even the 3D geometry of the objects is available.
Van Nguyen Nguyen+, Yuming Du+, Yang Xiao, Michaël Ramamonjisoa, Vincent Lepetit
3DV Oral, 2022 (+ equal contribution)
arXiv github

Star Fork

Templates for 3D Object Pose Estimation Revisited: Generalization to New Objects and Robustness to Occlusions
A method that can recognize objects and estimate their 3D pose in color images even under partial occlusions. Our method requires neither a training phase on these objects nor real images depicting them, only their CAD models.
Van Nguyen Nguyen, Yinlin Hu, Yang Xiao, Mathieu Salzmann, Vincent Lepetit
CVPR, 2022
project page   arxiv github

Star Fork


Talks
10/ 2022 CNOS: A Strong Baseline for CAD-based Novel Object Segmentation at ICCV 2023 R6D Workshop.

06/ 2022 PIZZA: A Powerful Image-only Zero-Shot Zero-CAD Approach to 6DoF Tracking at Oral presentation, 3DV 2022.

06/ 2022 Templates for 3D Object Pose Estimation Revisited at Poster Session, CVPR 2022.

12/ 2020 Continuous Implicit Functions (with Michaël Ramamonjisoa) at ENPC's seminar (slides).

Misc.
Reviewer: CVPR (2024 2023 2022), ICCV (2023), ECCV (2022 2024), 3DV (2024), TPAMI (2022), BMVC (2021).

Teaching: Image Processing and Artificial Vision (M1) at Ecole des Ponts ParisTech, 2021 & 2022.

Code: All of my released code is maintained on my GitHub account .

template-pose nope cnos pizza bop_viz_kit ObjectPoseSummary



Last update: 02/2024. This website takes the template from Jon Barron.