VIRTUS-FPP: Virtual Sensor Modeling for Fringe Projection Profilometry in NVIDIA Isaac Sim
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912609327906816 |
|---|---|
| author | Haroon, Adam Lakshman, Anush Balasubramaniam, Badrinath Li, Beiwen |
| author_facet | Haroon, Adam Lakshman, Anush Balasubramaniam, Badrinath Li, Beiwen |
| contents | Fringe projection profilometry (FPP) has been established as a high-accuracy 3D reconstruction method capable of achieving sub-pixel accuracy. However, this technique faces significant constraints due to complex calibration requirements, bulky system footprint, and sensitivity to environmental conditions. To address these limitations, we present VIRTUS-FPP, the first comprehensive physics-based virtual sensor modeling framework for FPP built in NVIDIA Isaac Sim. By leveraging the physics-based rendering and programmable sensing capabilities of simulation, our framework enables end-to-end modeling from calibration to reconstruction with full mathematical fidelity to the underlying principles of structured light. We conduct comprehensive virtual calibration and validate our system's reconstruction accuracy through quantitative comparison against ground truth geometry. Additionally, we demonstrate the ability to model the virtual system as a digital twin by replicating a physical FPP system in simulation and validating correspondence between virtual and real-world measurements. Experimental results demonstrate that VIRTUS-FPP accurately models optical phenomena critical to FPP and achieves results comparable to real-world systems while offering unprecedented flexibility for system configuration, sensor prototyping, and environmental control. This framework significantly accelerates the development of real-world FPP systems by enabling rapid virtual prototyping before physical implementation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_22685 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | VIRTUS-FPP: Virtual Sensor Modeling for Fringe Projection Profilometry in NVIDIA Isaac Sim Haroon, Adam Lakshman, Anush Balasubramaniam, Badrinath Li, Beiwen Image and Video Processing Computer Vision and Pattern Recognition Graphics I.3.7; I.4.1; I.4.5; I.6.3 Fringe projection profilometry (FPP) has been established as a high-accuracy 3D reconstruction method capable of achieving sub-pixel accuracy. However, this technique faces significant constraints due to complex calibration requirements, bulky system footprint, and sensitivity to environmental conditions. To address these limitations, we present VIRTUS-FPP, the first comprehensive physics-based virtual sensor modeling framework for FPP built in NVIDIA Isaac Sim. By leveraging the physics-based rendering and programmable sensing capabilities of simulation, our framework enables end-to-end modeling from calibration to reconstruction with full mathematical fidelity to the underlying principles of structured light. We conduct comprehensive virtual calibration and validate our system's reconstruction accuracy through quantitative comparison against ground truth geometry. Additionally, we demonstrate the ability to model the virtual system as a digital twin by replicating a physical FPP system in simulation and validating correspondence between virtual and real-world measurements. Experimental results demonstrate that VIRTUS-FPP accurately models optical phenomena critical to FPP and achieves results comparable to real-world systems while offering unprecedented flexibility for system configuration, sensor prototyping, and environmental control. This framework significantly accelerates the development of real-world FPP systems by enabling rapid virtual prototyping before physical implementation. |
| title | VIRTUS-FPP: Virtual Sensor Modeling for Fringe Projection Profilometry in NVIDIA Isaac Sim |
| topic | Image and Video Processing Computer Vision and Pattern Recognition Graphics I.3.7; I.4.1; I.4.5; I.6.3 |
| url | https://arxiv.org/abs/2509.22685 |