InvNeRF-Seg: Fine-Tuning a Pre-Trained NeRF for 3D Object Segmentation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910906335625216 |
|---|---|
| author | Zhao, Jiangsan Geipel, Jakob Kusnierek, Krzysztof Cui, Xuean |
| author_facet | Zhao, Jiangsan Geipel, Jakob Kusnierek, Krzysztof Cui, Xuean |
| contents | Neural Radiance Fields (NeRF) have been widely adopted for reconstructing high quality 3D point clouds from 2D RGB images. However, the segmentation of these reconstructed 3D scenes is more essential for downstream tasks such as object counting, size estimation, and scene understanding. While segmentation on raw 3D point clouds using deep learning requires labor intensive and time-consuming manual annotation, directly training NeRF on binary masks also fails due to the absence of color and shading cues essential for geometry learning. We propose Invariant NeRF for Segmentation (InvNeRFSeg), a two step, zero change fine tuning strategy for 3D segmentation. We first train a standard NeRF on RGB images and then fine tune it using 2D segmentation masks without altering either the model architecture or loss function. This approach produces higher quality, cleaner segmented point clouds directly from the refined radiance field with minimal computational overhead or complexity. Field density analysis reveals consistent semantic refinement: densities of object regions increase while background densities are suppressed, ensuring clean and interpretable segmentations. We demonstrate InvNeRFSegs superior performance over both SA3D and FruitNeRF on both synthetic fruit and real world soybean datasets. This approach effectively extends 2D segmentation to high quality 3D segmentation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_05751 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | InvNeRF-Seg: Fine-Tuning a Pre-Trained NeRF for 3D Object Segmentation Zhao, Jiangsan Geipel, Jakob Kusnierek, Krzysztof Cui, Xuean Computer Vision and Pattern Recognition Neural Radiance Fields (NeRF) have been widely adopted for reconstructing high quality 3D point clouds from 2D RGB images. However, the segmentation of these reconstructed 3D scenes is more essential for downstream tasks such as object counting, size estimation, and scene understanding. While segmentation on raw 3D point clouds using deep learning requires labor intensive and time-consuming manual annotation, directly training NeRF on binary masks also fails due to the absence of color and shading cues essential for geometry learning. We propose Invariant NeRF for Segmentation (InvNeRFSeg), a two step, zero change fine tuning strategy for 3D segmentation. We first train a standard NeRF on RGB images and then fine tune it using 2D segmentation masks without altering either the model architecture or loss function. This approach produces higher quality, cleaner segmented point clouds directly from the refined radiance field with minimal computational overhead or complexity. Field density analysis reveals consistent semantic refinement: densities of object regions increase while background densities are suppressed, ensuring clean and interpretable segmentations. We demonstrate InvNeRFSegs superior performance over both SA3D and FruitNeRF on both synthetic fruit and real world soybean datasets. This approach effectively extends 2D segmentation to high quality 3D segmentation. |
| title | InvNeRF-Seg: Fine-Tuning a Pre-Trained NeRF for 3D Object Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.05751 |