Gen-NeRF: Efficient and Generalizable Neural Radiance Fields via Algorithm-Hardware Co-Design

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Fu, Yonggan, Ye, Zhifan, Yuan, Jiayi, Zhang, Shunyao, Li, Sixu, You, Haoran, Lin, Yingyan Celine
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916550304333824
author Fu, Yonggan
Ye, Zhifan
Yuan, Jiayi
Zhang, Shunyao
Li, Sixu
You, Haoran
Lin, Yingyan Celine
author_facet Fu, Yonggan
Ye, Zhifan
Yuan, Jiayi
Zhang, Shunyao
Li, Sixu
You, Haoran
Lin, Yingyan Celine
contents Novel view synthesis is an essential functionality for enabling immersive experiences in various Augmented- and Virtual-Reality (AR/VR) applications, for which generalizable Neural Radiance Fields (NeRFs) have gained increasing popularity thanks to their cross-scene generalization capability. Despite their promise, the real-device deployment of generalizable NeRFs is bottlenecked by their prohibitive complexity due to the required massive memory accesses to acquire scene features, causing their ray marching process to be memory-bounded. To this end, we propose Gen-NeRF, an algorithm-hardware co-design framework dedicated to generalizable NeRF acceleration, which for the first time enables real-time generalizable NeRFs. On the algorithm side, Gen-NeRF integrates a coarse-then-focus sampling strategy, leveraging the fact that different regions of a 3D scene contribute differently to the rendered pixel, to enable sparse yet effective sampling. On the hardware side, Gen-NeRF highlights an accelerator micro-architecture to maximize the data reuse opportunities among different rays by making use of their epipolar geometric relationship. Furthermore, our Gen-NeRF accelerator features a customized dataflow to enhance data locality during point-to-hardware mapping and an optimized scene feature storage strategy to minimize memory bank conflicts. Extensive experiments validate the effectiveness of our proposed Gen-NeRF framework in enabling real-time and generalizable novel view synthesis.
format Preprint
id arxiv_https___arxiv_org_abs_2304_11842
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Gen-NeRF: Efficient and Generalizable Neural Radiance Fields via Algorithm-Hardware Co-Design
Fu, Yonggan
Ye, Zhifan
Yuan, Jiayi
Zhang, Shunyao
Li, Sixu
You, Haoran
Lin, Yingyan Celine
Computer Vision and Pattern Recognition
Hardware Architecture
Novel view synthesis is an essential functionality for enabling immersive experiences in various Augmented- and Virtual-Reality (AR/VR) applications, for which generalizable Neural Radiance Fields (NeRFs) have gained increasing popularity thanks to their cross-scene generalization capability. Despite their promise, the real-device deployment of generalizable NeRFs is bottlenecked by their prohibitive complexity due to the required massive memory accesses to acquire scene features, causing their ray marching process to be memory-bounded. To this end, we propose Gen-NeRF, an algorithm-hardware co-design framework dedicated to generalizable NeRF acceleration, which for the first time enables real-time generalizable NeRFs. On the algorithm side, Gen-NeRF integrates a coarse-then-focus sampling strategy, leveraging the fact that different regions of a 3D scene contribute differently to the rendered pixel, to enable sparse yet effective sampling. On the hardware side, Gen-NeRF highlights an accelerator micro-architecture to maximize the data reuse opportunities among different rays by making use of their epipolar geometric relationship. Furthermore, our Gen-NeRF accelerator features a customized dataflow to enhance data locality during point-to-hardware mapping and an optimized scene feature storage strategy to minimize memory bank conflicts. Extensive experiments validate the effectiveness of our proposed Gen-NeRF framework in enabling real-time and generalizable novel view synthesis.
title Gen-NeRF: Efficient and Generalizable Neural Radiance Fields via Algorithm-Hardware Co-Design
topic Computer Vision and Pattern Recognition
Hardware Architecture
url https://arxiv.org/abs/2304.11842