PFGS: High Fidelity Point Cloud Rendering via Feature Splatting

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Jiaxu, Zhang, Ziyi, He, Junhao, Xu, Renjing
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917713376444416
author Wang, Jiaxu
Zhang, Ziyi
He, Junhao
Xu, Renjing
author_facet Wang, Jiaxu
Zhang, Ziyi
He, Junhao
Xu, Renjing
contents Rendering high-fidelity images from sparse point clouds is still challenging. Existing learning-based approaches suffer from either hole artifacts, missing details, or expensive computations. In this paper, we propose a novel framework to render high-quality images from sparse points. This method first attempts to bridge the 3D Gaussian Splatting and point cloud rendering, which includes several cascaded modules. We first use a regressor to estimate Gaussian properties in a point-wise manner, the estimated properties are used to rasterize neural feature descriptors into 2D planes which are extracted from a multiscale extractor. The projected feature volume is gradually decoded toward the final prediction via a multiscale and progressive decoder. The whole pipeline experiences a two-stage training and is driven by our well-designed progressive and multiscale reconstruction loss. Experiments on different benchmarks show the superiority of our method in terms of rendering qualities and the necessities of our main components.
format Preprint
id arxiv_https___arxiv_org_abs_2407_03857
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle PFGS: High Fidelity Point Cloud Rendering via Feature Splatting
Wang, Jiaxu
Zhang, Ziyi
He, Junhao
Xu, Renjing
Computer Vision and Pattern Recognition
Rendering high-fidelity images from sparse point clouds is still challenging. Existing learning-based approaches suffer from either hole artifacts, missing details, or expensive computations. In this paper, we propose a novel framework to render high-quality images from sparse points. This method first attempts to bridge the 3D Gaussian Splatting and point cloud rendering, which includes several cascaded modules. We first use a regressor to estimate Gaussian properties in a point-wise manner, the estimated properties are used to rasterize neural feature descriptors into 2D planes which are extracted from a multiscale extractor. The projected feature volume is gradually decoded toward the final prediction via a multiscale and progressive decoder. The whole pipeline experiences a two-stage training and is driven by our well-designed progressive and multiscale reconstruction loss. Experiments on different benchmarks show the superiority of our method in terms of rendering qualities and the necessities of our main components.
title PFGS: High Fidelity Point Cloud Rendering via Feature Splatting
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.03857