학술논문

Plenoxels: Radiance Fields without Neural Networks
Document Type
Conference
Source
2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) CVPR Computer Vision and Pattern Recognition (CVPR), 2022 IEEE/CVF Conference on. :5491-5500 Jun, 2022
Subject
Computing and Processing
Visualization
Gradient methods
Computer vision
Three-dimensional displays
Codes
Neural networks
Benchmark testing
3D from multi-view and sensors; Efficient learning and inferences; Optimization methods; Vision + graphics
Language
ISSN
2575-7075
Abstract
We introduce Plenoxels (plenoptic voxels), a systemfor photorealistic view synthesis. Plenoxels represent a scene as a sparse 3D grid with spherical harmonics. This representation can be optimized from calibrated images via gradient methods and regularization without any neural components. On standard, benchmark tasks, Plenoxels are optimized two orders of magnitude faster than Neural Radiance Fields with no loss in visual quality. For video and code, please see https://alexyu.net/plenoxels.