SplatSDF: Boosting SDF-NeRF via Architecture-Level Fusion with Gaussian Splats

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Li, Runfa Blark, Suzuki, Keito, Du, Bang, Lee, Ki Myung Brian, Atanasov, Nikolay, Nguyen, Truong
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910033216798720
author Li, Runfa Blark
Suzuki, Keito
Du, Bang
Lee, Ki Myung Brian
Atanasov, Nikolay
Nguyen, Truong
author_facet Li, Runfa Blark
Suzuki, Keito
Du, Bang
Lee, Ki Myung Brian
Atanasov, Nikolay
Nguyen, Truong
contents Signed distance-radiance field (SDF-NeRF) is a promising environment representation that offers both photo-realistic rendering and geometric reasoning such as proximity queries for collision avoidance. However, the slow training speed and convergence of SDF-NeRF hinder their use in practical robotic systems. We propose SplatSDF, a novel SDF-NeRF architecture that accelerates convergence using 3D Gaussian splats (3DGS), which can be quickly pre-trained. Unlike prior approaches that introduce a consistency loss between separate 3DGS and SDF-NeRF models, SplatSDF directly fuses 3DGS at an architectural level by consuming it as an input to SDF-NeRF during training. This is achieved using a novel sparse 3DGS fusion strategy that injects neural embeddings of 3DGS into SDF-NeRF around the object surface, while also permitting inference without 3DGS for minimal operation. Experimental results show SplatSDF achieves 3X faster convergence to the same geometric accuracy than the best baseline, and outperforms state-of-the-art SDF-NeRF methods in terms of chamfer distance and peak signal to noise ratio, unlike consistency loss-based approaches that in fact provide limited gains. We also present computational techniques for accelerating gradient and Hessian steps by 3X. We expect these improvements will contribute to deploying SDF-NeRF on practical systems.
format Preprint
id arxiv_https___arxiv_org_abs_2411_15468
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SplatSDF: Boosting SDF-NeRF via Architecture-Level Fusion with Gaussian Splats
Li, Runfa Blark
Suzuki, Keito
Du, Bang
Lee, Ki Myung Brian
Atanasov, Nikolay
Nguyen, Truong
Computer Vision and Pattern Recognition
Graphics
Robotics
Signed distance-radiance field (SDF-NeRF) is a promising environment representation that offers both photo-realistic rendering and geometric reasoning such as proximity queries for collision avoidance. However, the slow training speed and convergence of SDF-NeRF hinder their use in practical robotic systems. We propose SplatSDF, a novel SDF-NeRF architecture that accelerates convergence using 3D Gaussian splats (3DGS), which can be quickly pre-trained. Unlike prior approaches that introduce a consistency loss between separate 3DGS and SDF-NeRF models, SplatSDF directly fuses 3DGS at an architectural level by consuming it as an input to SDF-NeRF during training. This is achieved using a novel sparse 3DGS fusion strategy that injects neural embeddings of 3DGS into SDF-NeRF around the object surface, while also permitting inference without 3DGS for minimal operation. Experimental results show SplatSDF achieves 3X faster convergence to the same geometric accuracy than the best baseline, and outperforms state-of-the-art SDF-NeRF methods in terms of chamfer distance and peak signal to noise ratio, unlike consistency loss-based approaches that in fact provide limited gains. We also present computational techniques for accelerating gradient and Hessian steps by 3X. We expect these improvements will contribute to deploying SDF-NeRF on practical systems.
title SplatSDF: Boosting SDF-NeRF via Architecture-Level Fusion with Gaussian Splats
topic Computer Vision and Pattern Recognition
Graphics
Robotics
url https://arxiv.org/abs/2411.15468