Evaluating geometric accuracy of NeRF reconstructions compared to SLAM method
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913446587531264 |
|---|---|
| author | Korycki, Adam Josephson, Colleen McGuire, Steve |
| author_facet | Korycki, Adam Josephson, Colleen McGuire, Steve |
| contents | As Neural Radiance Field (NeRF) implementations become faster, more efficient and accurate, their applicability to real world mapping tasks becomes more accessible. Traditionally, 3D mapping, or scene reconstruction, has relied on expensive LiDAR sensing. Photogrammetry can perform image-based 3D reconstruction but is computationally expensive and requires extremely dense image representation to recover complex geometry and photorealism. NeRFs perform 3D scene reconstruction by training a neural network on sparse image and pose data, achieving superior results to photogrammetry with less input data. This paper presents an evaluation of two NeRF scene reconstructions for the purpose of estimating the diameter of a vertical PVC cylinder. One of these are trained on commodity iPhone data and the other is trained on robot-sourced imagery and poses. This neural-geometry is compared to state-of-the-art lidar-inertial SLAM in terms of scene noise and metric-accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_11238 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Evaluating geometric accuracy of NeRF reconstructions compared to SLAM method Korycki, Adam Josephson, Colleen McGuire, Steve Computer Vision and Pattern Recognition Robotics As Neural Radiance Field (NeRF) implementations become faster, more efficient and accurate, their applicability to real world mapping tasks becomes more accessible. Traditionally, 3D mapping, or scene reconstruction, has relied on expensive LiDAR sensing. Photogrammetry can perform image-based 3D reconstruction but is computationally expensive and requires extremely dense image representation to recover complex geometry and photorealism. NeRFs perform 3D scene reconstruction by training a neural network on sparse image and pose data, achieving superior results to photogrammetry with less input data. This paper presents an evaluation of two NeRF scene reconstructions for the purpose of estimating the diameter of a vertical PVC cylinder. One of these are trained on commodity iPhone data and the other is trained on robot-sourced imagery and poses. This neural-geometry is compared to state-of-the-art lidar-inertial SLAM in terms of scene noise and metric-accuracy. |
| title | Evaluating geometric accuracy of NeRF reconstructions compared to SLAM method |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2407.11238 |