Few-Shot NeRF by Adaptive Rendering Loss Regularization | SpringerLink
Skip to main content
Springer Nature Link
Log in

Few-Shot NeRF by Adaptive Rendering Loss Regularization

  • Conference paper
  • First Online:
Computer Vision – ECCV 2024 (ECCV 2024)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 15124))

Included in the following conference series:

  • 213 Accesses

Abstract

Novel view synthesis with sparse inputs poses great challenges to Neural Radiance Field (NeRF). Recent works demonstrate that the frequency regularization of Positional Encoding (PE) can achieve promising results for few-shot NeRF. In this work, we reveal that there exists an inconsistency between the frequency regularization of PE and rendering loss. This prevents few-shot NeRF from synthesizing higher-quality novel views. To mitigate this inconsistency, we propose Adaptive Rendering loss regularization for few-shot NeRF, dubbed AR-NeRF. Specifically, we present a two-phase rendering supervision and an adaptive rendering loss weight learning strategy to align the frequency relationship between PE and 2D-pixel supervision. In this way, AR-NeRF can learn global structures better in the early training phase and adaptively learn local details throughout the training process. Extensive experiments show that our AR-NeRF achieves state-of-the-art performance on different datasets, including object-level and complex scenes. Our code will be available at https://github.com/GhiXu/AR-NeRF.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Subscribe and save

Springer+ Basic
¥17,985 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
JPY 3498
Price includes VAT (Japan)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
JPY 8007
Price includes VAT (Japan)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
JPY 10009
Price includes VAT (Japan)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Notes

  1. 1.

    A ray is a line emitted from a camera towards a pixel in that camera’s 2D view.

  2. 2.

    MixNeRF uses AlexNet to evaluate LPIPS while other methods use VGG. For fair comparisons, we use its provided models to retest its performance by VGG.

  3. 3.

    We retest the performance of FreeNeRF on DTU and LLFF by its provided models, and find that the results are slightly worse than the values reported in its paper.

References

  1. Bai, J., Huang, L., Gong, W., Guo, J., Guo, Y.: Self-nerf: a self-training pipeline for few-shot neural radiance fields. arXiv preprint arXiv:2303.05775 (2023)

  2. Barron, J.T., Mildenhall, B., Tancik, M., Hedman, P., Martin-Brualla, R., Srinivasan, P.P.: Mip-NeRF: a multiscale representation for anti-aliasing neural radiance fields. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 5855–5864 (2021)

    Google Scholar 

  3. Barron, J.T., Mildenhall, B., Verbin, D., Srinivasan, P.P., Hedman, P.: Mip-NeRF 360: unbounded anti-aliased neural radiance fields. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5470–5479 (2022)

    Google Scholar 

  4. Bradbury, J., et al.: Jax: composable transformations of python+ NumPy programs. Version 0.2 5, 14–24 (2018)

    Google Scholar 

  5. Chan, E.R., Monteiro, M., Kellnhofer, P., Wu, J., Wetzstein, G.: pi-GAN: periodic implicit generative adversarial networks for 3D-aware image synthesis. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5799–5809 (2021)

    Google Scholar 

  6. Chen, A., et al.: MVSNeRF: fast generalizable radiance field reconstruction from multi-view stereo. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 14124–14133 (2021)

    Google Scholar 

  7. Chen, D., Liu, Y., Huang, L., Wang, B., Pan, P.: GeoAug: data augmentation for few-shot nerf with geometry constraints. In: Avidan, S., Brostow, G., Cissé, M., Farinella, G.M., Hassner, T. (eds.) ECCV 2022. LNCS, vol. 13677, pp. 322–337. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-19790-1_20

    Chapter  Google Scholar 

  8. Chibane, J., Bansal, A., Lazova, V., Pons-Moll, G.: Stereo radiance fields (SRF): learning view synthesis for sparse views of novel scenes. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 7911–7920 (2021)

    Google Scholar 

  9. Deng, K., Liu, A., Zhu, J.Y., Ramanan, D.: Depth-supervised NeRF: fewer views and faster training for free. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 12882–12891 (2022)

    Google Scholar 

  10. Esposito, S., et al.: GeoGen: geometry-aware generative modeling via signed distance functions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 7479–7488 (2024)

    Google Scholar 

  11. Fu, Q., Xu, Q., Ong, Y.S., Tao, W.: Geo-Neus: geometry-consistent neural implicit surfaces learning for multi-view reconstruction. In: Advances in Neural Information Processing Systems, vol. 35, pp. 3403–3416 (2022)

    Google Scholar 

  12. Jain, A., Tancik, M., Abbeel, P.: Putting NeRF on a diet: semantically consistent few-shot view synthesis. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 5885–5894 (2021)

    Google Scholar 

  13. Jensen, R., Dahl, A., Vogiatzis, G., Tola, E., Aanæs, H.: Large scale multi-view stereopsis evaluation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 406–413 (2014)

    Google Scholar 

  14. Johari, M.M., Lepoittevin, Y., Fleuret, F.: GeoNeRF: generalizing nerf with geometry priors. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 18365–18375 (2022)

    Google Scholar 

  15. Kendall, A., Gal, Y.: What uncertainties do we need in Bayesian deep learning for computer vision? In: Advances in Neural Information Processing Systems, vol. 30 (2017)

    Google Scholar 

  16. Kim, M., Seo, S., Han, B.: InfoNeRF: ray entropy minimization for few-shot neural volume rendering. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 12912–12921 (2022)

    Google Scholar 

  17. Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

  18. Kwak, M., Song, J., Kim, S.: GecoNeRF: few-shot neural radiance fields via geometric consistency. arXiv preprint arXiv:2301.10941 (2023)

  19. Lin, C.H., et al.: Magic3D: high-resolution text-to-3D content creation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 300–309 (2023)

    Google Scholar 

  20. Martin-Brualla, R., Radwan, N., Sajjadi, M.S., Barron, J.T., Dosovitskiy, A., Duckworth, D.: Nerf in the wild: neural radiance fields for unconstrained photo collections. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 7210–7219 (2021)

    Google Scholar 

  21. Max, N.: Optical models for direct volume rendering. IEEE Trans. Visual Comput. Graph. 1(2), 99–108 (1995)

    Article  Google Scholar 

  22. Mildenhall, B., et al.: Local light field fusion: practical view synthesis with prescriptive sampling guidelines. ACM Trans. Graph. (TOG) 38(4), 1–14 (2019)

    Article  Google Scholar 

  23. Mildenhall, B., Srinivasan, P.P., Tancik, M., Barron, J.T., Ramamoorthi, R., Ng, R.: NeRF: representing scenes as neural radiance fields for view synthesis. Commun. ACM 65(1), 99–106 (2021)

    Article  Google Scholar 

  24. Niemeyer, M., Barron, J.T., Mildenhall, B., Sajjadi, M.S., Geiger, A., Radwan, N.: RegNeRF: regularizing neural radiance fields for view synthesis from sparse inputs. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5480–5490 (2022)

    Google Scholar 

  25. Pan, X., Lai, Z., Song, S., Huang, G.: ActiveNeRF: learning where to see with uncertainty estimation. In: Avidan, S., Brostow, G., Cissé, M., Farinella, G.M., Hassner, T. (eds.) ECCV 2022. LNCS, vol. 13693, pp. 230–246. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-19827-4_14

    Chapter  Google Scholar 

  26. Poole, B., Jain, A., Barron, J.T., Mildenhall, B.: DreamFusion: text-to-3D using 2D diffusion. arXiv preprint arXiv:2209.14988 (2022)

  27. Ren, C., Xu, Q., Zhang, S., Yang, J.: Hierarchical prior mining for non-local multi-view stereo. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 3611–3620 (2023)

    Google Scholar 

  28. Roessle, B., Barron, J.T., Mildenhall, B., Srinivasan, P.P., Nießner, M.: Dense depth priors for neural radiance fields from sparse input views. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 12892–12901 (2022)

    Google Scholar 

  29. Seo, S., Chang, Y., Kwak, N.: FlipNeRF: flipped reflection rays for few-shot novel view synthesis. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 22883–22893 (2023)

    Google Scholar 

  30. Seo, S., Han, D., Chang, Y., Kwak, N.: MixNeRF: modeling a ray with mixture density for novel view synthesis from sparse inputs. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 20659–20668 (2023)

    Google Scholar 

  31. Shen, J., Agudo, A., Moreno-Noguer, F., Ruiz, A.: Conditional-flow nerf: accurate 3D modelling with reliable uncertainty quantification. In: Avidan, S., Brostow, G., Cissé, M., Farinella, G.M., Hassner, T. (eds.) ECCV 2022. LNCS, vol. 13663, pp. 540–557. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-20062-5_31

    Chapter  Google Scholar 

  32. Shen, J., Ruiz, A., Agudo, A., Moreno-Noguer, F.: Stochastic neural radiance fields: quantifying uncertainty in implicit 3D representations. In: Proceedings of the International Conference on 3D Vision, pp. 972–981. IEEE (2021)

    Google Scholar 

  33. Su, W., Zhang, C., Xu, Q., Tao, W.: PSDF: prior-driven neural implicit surface learning for multi-view reconstruction. arXiv preprint arXiv:2401.12751 (2024)

  34. Sun, X., Xu, Q., Yang, X., Zang, Y., Wang, C.: Global and hierarchical geometry consistency priors for few-shot nerfs in indoor scenes. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 20530–20539 (2024)

    Google Scholar 

  35. Tancik, M., et al.: Fourier features let networks learn high frequency functions in low dimensional domains. In: Advances in Neural Information Processing Systems, vol. 33, pp. 7537–7547 (2020)

    Google Scholar 

  36. Truong, P., Rakotosaona, M.J., Manhardt, F., Tombari, F.: SPARF: neural radiance fields from sparse and noisy poses. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4190–4200 (2023)

    Google Scholar 

  37. Uy, M.A., Martin-Brualla, R., Guibas, L., Li, K.: SCADE: NeRFs from space carving with ambiguity-aware depth estimates. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 16518–16527 (2023)

    Google Scholar 

  38. Verbin, D., Hedman, P., Mildenhall, B., Zickler, T., Barron, J.T., Srinivasan, P.P.: Ref-NeRF: structured view-dependent appearance for neural radiance fields. In: 2022 IEEE Conference on Computer Vision and Pattern Recognition, pp. 5481–5490. IEEE (2022)

    Google Scholar 

  39. Wang, G., Chen, Z., Loy, C.C., Liu, Z.: SparseNeRF: distilling depth ranking for few-shot novel view synthesis. arXiv preprint arXiv:2303.16196 (2023)

  40. Wang, H., Xu, Q., Chen, H., Ma, R.: PGAHum: prior-guided geometry and appearance learning for high-fidelity animatable human reconstruction. arXiv preprint arXiv:2404.13862 (2024)

  41. Wang, H., Du, X., Li, J., Yeh, R.A., Shakhnarovich, G.: Score Jacobian chaining: lifting pretrained 2D diffusion models for 3d generation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 12619–12629 (2023)

    Google Scholar 

  42. Wang, P., Liu, L., Liu, Y., Theobalt, C., Komura, T., Wang, W.: NeuS: learning neural implicit surfaces by volume rendering for multi-view reconstruction. In: Advances in Neural Information Processing Systems (2021)

    Google Scholar 

  43. Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)

    Article  Google Scholar 

  44. Wu, R., et al.: Reconfusion: 3D reconstruction with diffusion priors. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 21551–21561 (2024)

    Google Scholar 

  45. Wynn, J., Turmukhambetov, D.: DiffusioNeRF: regularizing neural radiance fields with denoising diffusion models. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4180–4189 (2023)

    Google Scholar 

  46. Xu, J., et al.: PR-Neus: a prior-based residual learning paradigm for fast multi-view neural surface reconstruction. arXiv preprint arXiv:2312.11577 (2023)

  47. Xu, Q., Kong, W., Tao, W., Pollefeys, M.: Multi-scale geometric consistency guided and planar prior assisted multi-view stereo. IEEE Trans. Pattern Anal. Mach. Intell. 45(4), 4945–4963 (2022)

    Google Scholar 

  48. Xu, Q., Tao, W.: Multi-scale geometric consistency guided multi-view stereo. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5483–5492 (2019)

    Google Scholar 

  49. Yang, J., Pavone, M., Wang, Y.: FreeNeRF: improving few-shot neural rendering with free frequency regularization. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 8254–8263 (2023)

    Google Scholar 

  50. Yariv, L., Gu, J., Kasten, Y., Lipman, Y.: Volume rendering of neural implicit surfaces. In: Advances in Neural Information Processing Systems, vol. 34, pp. 4805–4815 (2021)

    Google Scholar 

  51. Yi, X., Wu, Z., Xu, Q., Zhou, P., Lim, J.H., Zhang, H.: Diffusion time-step curriculum for one image to 3D generation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 9948–9958 (2024)

    Google Scholar 

  52. Yu, A., Ye, V., Tancik, M., Kanazawa, A.: pixelNeRF: neural radiance fields from one or few images. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4578–4587 (2021)

    Google Scholar 

  53. Zhang, R., Isola, P., Efros, A.A., Shechtman, E., Wang, O.: The unreasonable effectiveness of deep features as a perceptual metric. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 586–595 (2018)

    Google Scholar 

Download references

Acknowledgements

This research is supported by the National Research Foundation, Singapore under its AI Singapore Programme (AISG Award No: AISG2-RP-2021-022). Xuanyu Yi is supported by the Agency for Science, Technology AND Research, Singapore.

Author information

Authors and Affiliations

  1. Nanyang Technological University, Singapore, Singapore

    Qingshan Xu, Xuanyu Yi, Yew-Soon Ong & Hanwang Zhang

  2. Huazhong University of Science and Technology, Wuhan, China

    Jianyao Xu & Wenbing Tao

  3. A*STAR, Singapore, Singapore

    Yew-Soon Ong

Authors
  1. Qingshan Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Xuanyu Yi
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jianyao Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Wenbing Tao
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Yew-Soon Ong
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Hanwang Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Qingshan Xu .

Editor information

Editors and Affiliations

  1. University of Birmingham, Birmingham, UK

    Aleš Leonardis

  2. University of Trento, Trento, Italy

    Elisa Ricci

  3. Technical University of Darmstadt, Darmstadt, Germany

    Stefan Roth

  4. Princeton University, Princeton, NJ, USA

    Olga Russakovsky

  5. Czech Technical University in Prague, Prague, Czech Republic

    Torsten Sattler

  6. École des Ponts ParisTech, Marne-la-Vallée, France

    Gül Varol

1 Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 2093 KB)

Rights and permissions

Reprints and permissions

Copyright information

© 2025 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Xu, Q., Yi, X., Xu, J., Tao, W., Ong, YS., Zhang, H. (2025). Few-Shot NeRF by Adaptive Rendering Loss Regularization. In: Leonardis, A., Ricci, E., Roth, S., Russakovsky, O., Sattler, T., Varol, G. (eds) Computer Vision – ECCV 2024. ECCV 2024. Lecture Notes in Computer Science, vol 15124. Springer, Cham. https://doi.org/10.1007/978-3-031-72848-8_8

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-72848-8_8

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-72847-1

  • Online ISBN: 978-3-031-72848-8

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics