journal article·2025-05-12Open access

Frequency and tensorial neural radiance fields for digital preservation of ancient buildings

Chenxu Li, Dong Sui, Jin Li, Yacong Li, Yang Li, Maozu Guo

npj Heritage Science· Volume 13· Issue 1

Abstract

Ancient buildings are invaluable cultural heritage assets, traditional preservation methods often suffer from inefficiency and low accuracy. We propose FaTNeRF, a NeRF-based framework that integrates adaptive frequency encoding and a novel density loss for high-quality rendering and digital preservation. Evaluations on public datasets show our method achieves fast reconstruction (<20 min) with superior visual quality while maintaining compact model size (<75 MB). Additional validation on a self-collected cultural heritage dataset confirms its effectiveness in complex real-world scenarios. FaTNeRF advances digital preservation by enabling efficient, high-fidelity 3D reconstruction for historic buildings.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: 3D Surveying and Cultural Heritage · Conservation Techniques and Studies · Image Processing and 3D Reconstruction

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (45)

Show source-provided references
1. Karami, A., Menna, F. & Remondino, F. Combining photogrammetry and photometric stereo to achieve precise and complete 3d reconstruction. Sensors 22, 8172 (2022).10.3390/s22218172 ↗
2. Choe, J., Im, S., Rameau, F., Kang, M. & Kweon, I. S. VolumeFusion: deep depth fusion for 3D scene reconstruction. In Proc. IEEE/CVF International Conference on Computer Vision, 16086–16095 (IEEE, 2021).10.1109/ICCV48922.2021.01578 ↗
3. Zhou, L., Sun, X., Zhang, C., Cao, L. & Li, Y. Lidar-based 3d glass detection and reconstruction in indoor environment. IEEE Trans. Instrum. Meas. 73, 8502211 (IEEE, 2024).10.1109/TIM.2024.3375965 ↗
4. Hou, M., Huo, D., Yang, Y., Yang, S. & Chen, H. Using mask R-CNN to rapidly detect the gold foil shedding of stone cultural heritage in images. Heritage Sci. 12, 46 (2024).10.1186/s40494-024-01158-9 ↗
5. Zhu, J., Fang, B., Chen, T. & Yang, H. Face repairing based on transfer learning method with fewer training samples: application to a terracotta warrior with facial cracks and a buddha with a broken nose. Heritage Sci. 12, 186 (2024).10.1186/s40494-024-01292-4 ↗
6. Liu, Z. et al. Semantic segmentation and photogrammetry of crowdsourced images to monitor historic facades. Heritage Sci. 10, 1–17 (2022).10.1186/s40494-022-00664-y ↗
7. Yao, X., Hu, L. & Zhang, J. Reconstructing the local structures of chinese ancient architecture using unsupervised depth estimation. Heritage Sci. 12, 318 (2024).10.1186/s40494-024-01433-9 ↗
8. Mildenhall, B. et al. Nerf: Representing scenes as neural radiance fields for view synthesis. Commun. ACM 65, 99–106 (2021).10.1145/3503250 ↗
9. Molaei, A. et al. Implicit neural representation in medical imaging: a comparative survey. In Proc. IEEE/CVF International Conference on Computer Vision, 2381–2391 (IEEE, 2023).10.1109/ICCVW60793.2023.00252 ↗
10. Corona-Figueroa, A. et al. MedNeRF: Medical neural radiance fields for reconstructing 3d-aware VT-projections from a single x-ray. In Proc. 44th annual international conference of the IEEE engineering in medicine & Biology society (EMBC), 3843–3848 (IEEE, 2022).10.1109/EMBC48229.2022.9871757 ↗
11. He, L. et al. Neural radiance field in autonomous driving: a survey. arXiv preprint arXiv:2404.13816 (2024).
12. Cao, J. et al. Lightning NeRF: Efficient hybrid scene representation for autonomous driving. In Proc. IEEE International Conference on Robotics and Automation, 16803–16809 (IEEE, 2024).10.1109/ICRA57147.2024.10611130 ↗
13. Tonderski, A. et al. NeuRAD: Neural rendering for autonomous driving. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 14895–14904 (IEEE, 2024).10.1109/CVPR52733.2024.01411 ↗
14. Du, Y., Zhang, Y., Yu, H.-X., Tenenbaum, J. B. & Wu, J. Neural radiance flow for 4d view synthesis and video processing. In Proc. IEEE/CVF International Conference on Computer Vision (ICCV), 14304–14314 (IEEE Computer Society, 2021).10.1109/ICCV48922.2021.01406 ↗
15. Turki, H., Ramanan, D. & Satyanarayanan, M. Mega-NeRF: Scalable construction of large-scale nerfs for virtual fly-throughs. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12922–12931 (IEEE, 2022).10.1109/CVPR52688.2022.01258 ↗
16. Tretschk, E. et al. Non-rigid neural radiance fields: Reconstruction and novel view synthesis of a dynamic scene from monocular video. In Proc. IEEE/CVF International Conference on Computer Vision, 12959–12970 (IEEE, 2021).10.1109/ICCV48922.2021.01272 ↗
17. Wang, C., Chai, M., He, M., Chen, D. & Liao, J. Clip-NeRF: Text-and-image driven manipulation of neural radiance fields. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 3835–3844 (IEEE, 2022).10.1109/CVPR52688.2022.00381 ↗
18. Hu, B., Huang, J., Liu, Y., Tai, Y.-W. & Tang, C.-K. NeRF-RPN: A general framework for object detection in nerfs. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 23528–23538 (IEEE, 2023).10.1109/CVPR52729.2023.02253 ↗
19. Adamkiewicz, M. et al. Vision-only robot navigation in a neural radiance world. IEEE Robot. Automat. Lett. 7, 4606–4613 (2022).10.1109/LRA.2022.3150497 ↗
20. Tancik, M. et al. Block-NeRF: Scalable large scene neural view synthesis. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8248–8258 (IEEE, 2022).10.1109/CVPR52688.2022.00807 ↗
21. Pumarola, A., Corona, E., Pons-Moll, G. & Moreno-Noguer, F. D-NeRF: Neural radiance fields for dynamic scenes. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 10318–10327 (IEEE, 2021).10.1109/CVPR46437.2021.01018 ↗
22. Wang, Z., Wu, S., Xie, W., Chen, M. & Prisacariu, V. A. NeRF–: Neural radiance fields without known camera parameters. arXiv preprint arXiv:2102.07064 (2021).
23. Barron, J. T., Mildenhall, B., Verbin, D., Srinivasan, P. P. & Hedman, P. Zip-NeRF: Anti-aliased grid-based neural radiance fields. In Proc. IEEE/CVF International Conference on Computer Vision, 19697–19705 (IEEE, 2023).10.1109/ICCV51070.2023.01804 ↗
24. Barron, J. T., Mildenhall, B., Verbin, D., Srinivasan, P. P. & Hedman, P. Mip-NeRF 360: Unbounded anti-aliased neural radiance fields. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5470–5479 (IEEE, 2022).10.1109/CVPR52688.2022.00539 ↗
25. Müller, T., Evans, A., Schied, C. & Keller, A. Instant neural graphics primitives with a multiresolution hash encoding. ACM Trans. Graph. 41, 1–15 (2022).10.1145/3528223.3530127 ↗
26. Barron, J. T. et al. Mip-NeRF: A multiscale representation for anti-aliasing neural radiance fields. In Proc. IEEE/CVF International Conference on Computer Vision, 5855–5864 (IEEE, 2021).10.1109/ICCV48922.2021.00580 ↗
27. Yu, A. et al. Plenoctrees for real-time rendering of neural radiance fields. In Proc. IEEE/CVF International Conference on Computer Vision, 5752–5761 (IEEE, 2021).10.1109/ICCV48922.2021.00570 ↗
28. Park, K. et al. Nerfies: Deformable neural radiance fields. In Proc. IEEE/CVF International Conference on Computer Vision, 5865–5874 (IEEE, 2021).10.1109/ICCV48922.2021.00581 ↗
29. Xie, Z., Zhang, J., Li, W., Zhang, F. & Zhang, L. S-NeRF: Neural radiance fields for street views. arXiv preprint arXiv:2303.00749 (2023).
30. Reiser, C., Peng, S., Liao, Y. & Geiger, A. KiloNeRF: Speeding up neural radiance fields with thousands of tiny MLPs. In Proc. IEEE/CVF International Conference on Computer Vision, 14335–14345 (IEEE, 2021).10.1109/ICCV48922.2021.01407 ↗
31. Hu, T., Liu, S., Chen, Y., Shen, T. & Jia, J. Efficientnerf efficient neural radiance fields. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12902–12911 (IEEE, 2022).10.1109/CVPR52688.2022.01256 ↗
32. Fridovich-Keil, S. et al. Plenoxels: Radiance fields without neural networks. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5501–5510 (IEEE, 2022).10.1109/CVPR52688.2022.00542 ↗
33. Sun, C., Sun, M. & Chen, H.-T. Direct voxel grid optimization: Super-fast convergence for radiance fields reconstruction. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5459–5469 (IEEE, 2022).10.1109/CVPR52688.2022.00538 ↗
34. Chen, A., Xu, Z., Geiger, A., Yu, J. & Su, H. Tensorf: Tensorial radiance fields. In European conference on computer vision, 333–350 (Springer, 2022).10.1007/978-3-031-19824-3_20 ↗
35. Tang, J., Chen, X., Wang, J. & Zeng, G. Compressible-composable NeRF via rank-residual decomposition. Adv. Neural Inform. Process. Syste. 35, 14798–14809 (2022).
36. Yuan, S. & Zhao, H. Slimmerf: slimmable radiance fields. In 2024 International Conference on 3D Vision (3DV), 64–74 (IEEE, 2024).10.1109/3DV62453.2024.00143 ↗
37. Li, L., Shen, Z., Wang, Z., Shen, L. & Bo, L. Compressing volumetric radiance fields to 1 mb. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 4222–4231 (IEEE, 2023).10.1109/CVPR52729.2023.00411 ↗
38. Somraj, N. & Soundararajan, R. Vip-NeRF: Visibility prior for sparse input neural radiance fields. In ACM SIGGRAPH 2023 Conference Proceedings, 1–11 (ACM, 2023).10.1145/3610548.3618188 ↗
39. Deng, K., Liu, A., Zhu, J.-Y. & Ramanan, D. Depth-supervised NeRF: fewer views and faster training for free. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12882–12891 (IEEE, 2022).10.1109/CVPR52688.2022.01254 ↗
40. Wang, G., Chen, Z., Loy, C. C. & Liu, Z. SparseNeRF: distilling depth ranking for few-shot novel view synthesis. In Proc. IEEE/CVF International Conference on Computer Vision, 9065–9076 (IEEE, 2023).10.1109/ICCV51070.2023.00832 ↗
41. Jain, A., Tancik, M. & Abbeel, P. Putting NeRF on a diet: Semantically consistent few-shot view synthesis. In Proc. IEEE/CVF International Conference on Computer Vision, 5885–5894 (IEEE, 2021).10.1109/ICCV48922.2021.00583 ↗
42. Niemeyer, M. et al. RegNeRF: Regularizing neural radiance fields for view synthesis from sparse inputs. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5480–5490 (IEEE, 2022).10.1109/CVPR52688.2022.00540 ↗
43. Tancik, M. et al. Fourier features let networks learn high frequency functions in low dimensional domains. Adv. Neural Inform. Process. Syst. 33, 7537–7547 (2020).
44. Yang, J., Pavone, M. & Wang, Y. FreeNeRF: Improving few-shot neural rendering with free frequency regularization. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8254–8263 (IEEE, 2023).10.1109/CVPR52729.2023.00798 ↗
45. Paszke, A. et al. Pytorch: An imperative style, high-performance deep learning library. Adv. Neural inform. Process. Syst. 32, 8024–8035 Curran Associates, Inc., (2019).