Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2402.08138 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911791437578240 |
|---|---|
| author | Park, Minyoung Do, Mirae Shin, YeonJae Yoo, Jaeseok Hong, Jongkwang Kim, Joongrock Lee, Chul |
| author_facet | Park, Minyoung Do, Mirae Shin, YeonJae Yoo, Jaeseok Hong, Jongkwang Kim, Joongrock Lee, Chul |
| contents | Advanced techniques using Neural Radiance Fields (NeRF), Signed Distance Fields (SDF), and Occupancy Fields have recently emerged as solutions for 3D indoor scene reconstruction. We introduce a novel two-phase learning approach, H2O-SDF, that discriminates between object and non-object regions within indoor environments. This method achieves a nuanced balance, carefully preserving the geometric integrity of room layouts while also capturing intricate surface details of specific objects. A cornerstone of our two-phase learning framework is the introduction of the Object Surface Field (OSF), a novel concept designed to mitigate the persistent vanishing gradient problem that has previously hindered the capture of high-frequency details in other methods. Our proposed approach is validated through several experiments that include ablation studies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_08138 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | H2O-SDF: Two-phase Learning for 3D Indoor Reconstruction using Object Surface Fields Park, Minyoung Do, Mirae Shin, YeonJae Yoo, Jaeseok Hong, Jongkwang Kim, Joongrock Lee, Chul Computer Vision and Pattern Recognition Advanced techniques using Neural Radiance Fields (NeRF), Signed Distance Fields (SDF), and Occupancy Fields have recently emerged as solutions for 3D indoor scene reconstruction. We introduce a novel two-phase learning approach, H2O-SDF, that discriminates between object and non-object regions within indoor environments. This method achieves a nuanced balance, carefully preserving the geometric integrity of room layouts while also capturing intricate surface details of specific objects. A cornerstone of our two-phase learning framework is the introduction of the Object Surface Field (OSF), a novel concept designed to mitigate the persistent vanishing gradient problem that has previously hindered the capture of high-frequency details in other methods. Our proposed approach is validated through several experiments that include ablation studies. |
| title | H2O-SDF: Two-phase Learning for 3D Indoor Reconstruction using Object Surface Fields |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.08138 |