Full search page →

The multiresolution hash encoding

Optional · supplementary note 5 of 7 for DLCV 11 · 3D Reconstruction: From Multi-View Geometry to NeRF and Gaussian Splatting. Nothing in the main text depends on it; citation numbers refer to the reference list at the bottom of this page.

Navigate: this page and this track: Navigate: this page and this track

On this page

Supplementary notes · 7

  1. The normalized eight-point algorithm
  2. Linear triangulation (DLT)
  3. Bundle adjustment in code
  4. The plane-induced homography
  5. The multiresolution hash encoding
  6. Why NeuS weights peak on the surface
  7. Projecting a 3D Gaussian: the local affine approximation

Paper guides · 3

  1. NeRF (Mildenhall et al., 2020)
  2. 3D Gaussian Splatting (Kerbl et al., 2023)
  3. Guide to the Modern view surveys

Deep Learning for Computer Vision

  1. 02 · Deep Image Classification
View the whole course tree →

This note gives the details of the Instant-NGP encoding summarized in the main text: how the grid resolutions are chosen, how grid corners are hashed, and how many parameters the encoding has.

Levels. The encoding has LL levels. Level ll, for l=0,…,L−1l = 0, \dots, L - 1, is a grid of resolution

Nl=⌊Nmin⁡ b l⌋,b=exp⁡ ⁣(ln⁡Nmax⁡−ln⁡Nmin⁡L−1),N_l = \left\lfloor N_{\min} \, b^{\,l} \right\rfloor, \qquad b = \exp\!\left( \frac{\ln N_{\max} - \ln N_{\min}}{L - 1} \right),

where Nmin⁡N_{\min} and Nmax⁡N_{\max} are the coarsest and finest resolutions, ⌊⋅⌋\lfloor \cdot \rfloor rounds down, and bb is the growth factor, so the resolutions form a geometric progression [19]. The paper’s typical values are L=16L = 16 levels, F=2F = 2 features per table entry, TT between 2142^{14} and 2242^{24} entries per level, Nmin⁡=16N_{\min} = 16, and Nmax⁡N_{\max} between 512 and 524,288, chosen to match the finest detail in the data [19].

Lookup. On level ll, the input point is scaled by NlN_l, and the grid cell that contains it has 2d2^d corners with integer coordinates (d=3d = 3 in 3D). If the whole grid fits in the table, that is (Nl+1)d≤T(N_l + 1)^d \le T, every corner gets its own entry. Otherwise a corner x=(x1,…,xd)\mathbf{x} = (x_1, \dots, x_d) uses the entry

h(x)=(⨁i=1dxiπi) mod T,h(\mathbf{x}) = \left( \bigoplus_{i=1}^{d} x_i \pi_i \right) \bmod T,

where ⊕\oplus is bitwise XOR and the constants are π1=1\pi_1 = 1, π2=2 654 435 761\pi_2 = 2\,654\,435\,761 and π3=805 459 861\pi_3 = 805\,459\,861 [19]. Only d−1d - 1 of the dimensions need a large prime to decorrelate them, and π1=1\pi_1 = 1 keeps corners that are neighbours along the first axis next to each other in memory [19]. The features of the 2d2^d corners are interpolated linearly in each dimension (trilinearly in 3D) according to the point’s position inside the cell. The interpolated vectors of all LL levels are concatenated, together with any extra inputs such as the encoded viewing direction, and passed to the MLP [19].

Collisions. On fine levels many corners share one entry, and the encoding does not detect or resolve this. The gradients of colliding corners are averaged, so the corners with the largest gradients, those most relevant to the loss, dominate the shared entry; the authors argue that the table therefore concentrates on the sparse regions with the most important fine detail, and that the multiresolution structure lets the network tell collisions apart [19].

Parameter count. The tables hold at most T⋅L⋅FT \cdot L \cdot F trainable numbers, whatever Nmax⁡N_{\max} is [19]. Memory is fixed by TT, not by the finest resolution.

A worked example. The snippet computes the levels for Nmax⁡=2048N_{\max} = 2048 and T=219T = 2^{19}, values inside the paper’s typical ranges, and hashes three neighbouring corners.

Code · Python · 19 lines
import numpy as np

L, F, T = 16, 2, 2 ** 19                                 # levels, features per entry, table size
N_min, N_max = 16, 2048                                  # coarsest and finest grid resolution
b = np.exp((np.log(N_max) - np.log(N_min)) / (L - 1))   # growth factor between levels
N = np.floor(N_min * b ** np.arange(L)).astype(int)     # resolution of each level
dense = (N + 1) ** 3 <= T                                # does the full grid fit in the table?
print(f"growth factor b = {b:.3f}; resolutions {N.tolist()}")
print(f"levels stored 1:1 (no hashing): {dense.sum()}, hashed levels: {(~dense).sum()}")
print(f"finest level: {(N[-1] + 1) ** 3:,} grid vertices share {T:,} entries "
      f"(about {(N[-1] + 1) ** 3 / T:,.0f} per entry)")
print(f"trainable encoding parameters: at most T * L * F = {T * L * F:,}")

PI = np.array([1, 2654435761, 805459861], dtype=np.uint64)     # the primes of the paper (pi_1 = 1)
def spatial_hash(corner):
    """h(x) = (x_1 pi_1  XOR  x_2 pi_2  XOR  x_3 pi_3) mod T for an integer grid corner."""
    return int(np.bitwise_xor.reduce(np.asarray(corner, dtype=np.uint64) * PI) % np.uint64(T))
for corner in [(100, 200, 300), (101, 200, 300), (100, 201, 300)]:
    print(f"corner {corner} -> table entry {spatial_hash(corner)}")

Output

growth factor b = 1.382; resolutions [16, 22, 30, 42, 58, 80, 111, 153, 212, 294, 406, 561, 776, 1072, 1482, 2048]
levels stored 1:1 (no hashing): 5, hashed levels: 11
finest level: 8,602,523,649 grid vertices share 524,288 entries (about 16,408 per entry)
trainable encoding parameters: at most T * L * F = 16,777,216
corner (100, 200, 300) -> table entry 110768
corner (101, 200, 300) -> table entry 110769
corner (100, 201, 300) -> table entry 10497

With these settings the five coarsest levels fit their tables directly and the other eleven are hashed. On the finest level, about 16,000 grid corners share each table entry, which is workable only because most of them never receive a large gradient. The last three lines show the effect of π1=1\pi_1 = 1: a step along the first axis moves to the next entry, while a step along the second axis jumps far away.

References

  1. R. Hartley and A. Zisserman, Multiple View Geometry in Computer Vision, 2nd ed., Cambridge University Press, 2004. doi
  2. J. L. Schönberger and J.-M. Frahm, “Structure-from-motion revisited,” in Proc. IEEE CVPR, 2016. CVF
  3. J. L. Schönberger, E. Zheng, J.-M. Frahm and M. Pollefeys, “Pixelwise view selection for unstructured multi-view stereo,” in Proc. ECCV, 2016. doi
  4. J. L. Schönberger et al., “COLMAP,” official documentation (general-purpose structure-from-motion and multi-view stereo pipeline). link
  5. B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoorthi and R. Ng, “NeRF: Representing scenes as neural radiance fields for view synthesis,” in Proc. ECCV, 2020; also Communications of the ACM, vol. 65, no. 1, pp. 99–106, 2022. doi · CACM doi · arXiv
  6. B. Kerbl, G. Kopanas, T. Leimkühler and G. Drettakis, “3D Gaussian Splatting for real-time radiance field rendering,” ACM Transactions on Graphics, vol. 42, no. 4, 2023. doi · arXiv
  7. H. C. Longuet-Higgins, “A computer algorithm for reconstructing a scene from two projections,” Nature, vol. 293, pp. 133–135, 1981. doi
  8. R. I. Hartley, “In defense of the eight-point algorithm,” IEEE TPAMI, vol. 19, no. 6, pp. 580–593, 1997. doi
  9. D. Nistér, “An efficient solution to the five-point relative pose problem,” IEEE TPAMI, vol. 26, no. 6, pp. 756–770, 2004. doi
  10. M. A. Fischler and R. C. Bolles, “Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography,” Communications of the ACM, vol. 24, no. 6, pp. 381–395, 1981. doi
  11. D. G. Lowe, “Distinctive image features from scale-invariant keypoints,” International Journal of Computer Vision, vol. 60, no. 2, pp. 91–110, 2004. doi
  12. N. Snavely, S. M. Seitz and R. Szeliski, “Photo tourism: Exploring photo collections in 3D,” ACM Transactions on Graphics, vol. 25, no. 3, pp. 835–846, 2006. doi
  13. B. Triggs, P. F. McLauchlan, R. I. Hartley and A. W. Fitzgibbon, “Bundle adjustment — A modern synthesis,” in Vision Algorithms: Theory and Practice, LNCS 1883, pp. 298–372, Springer, 2000. doi
  14. R. T. Collins, “A space-sweep approach to true multi-image matching,” in Proc. IEEE CVPR, pp. 358–363, 1996. doi
  15. B. Curless and M. Levoy, “A volumetric method for building complex models from range images,” in Proc. SIGGRAPH, pp. 303–312, 1996. doi
  16. Y. Yao, Z. Luo, S. Li, T. Fang and L. Quan, “MVSNet: Depth inference for unstructured multi-view stereo,” in Proc. ECCV, 2018. CVF · doi
  17. N. Max, “Optical models for direct volume rendering,” IEEE Transactions on Visualization and Computer Graphics, vol. 1, no. 2, pp. 99–108, 1995. doi
  18. M. Tancik, P. P. Srinivasan, B. Mildenhall, S. Fridovich-Keil, N. Raghavan, U. Singhal, R. Ramamoorthi, J. T. Barron and R. Ng, “Fourier features let networks learn high frequency functions in low dimensional domains,” in Proc. NeurIPS, 2020. proceedings
  19. T. Müller, A. Evans, C. Schied and A. Keller, “Instant neural graphics primitives with a multiresolution hash encoding,” ACM Transactions on Graphics, vol. 41, no. 4, 2022. doi · arXiv
  20. J. J. Park, P. Florence, J. Straub, R. Newcombe and S. Lovegrove, “DeepSDF: Learning continuous signed distance functions for shape representation,” in Proc. IEEE/CVF CVPR, 2019. CVF
  21. L. Mescheder, M. Oechsle, M. Niemeyer, S. Nowozin and A. Geiger, “Occupancy networks: Learning 3D reconstruction in function space,” in Proc. IEEE/CVF CVPR, 2019. CVF
  22. P. Wang, L. Liu, Y. Liu, C. Theobalt, T. Komura and W. Wang, “NeuS: Learning neural implicit surfaces by volume rendering for multi-view reconstruction,” in Proc. NeurIPS, 2021. proceedings · arXiv
  23. M. Zwicker, H. Pfister, J. van Baar and M. Gross, “EWA volume splatting,” in Proc. IEEE Visualization, 2001. doi
  24. W. E. Lorensen and H. E. Cline, “Marching cubes: A high resolution 3D surface construction algorithm,” in Proc. SIGGRAPH, pp. 163–169, 1987. doi
  25. M. Kazhdan, M. Bolitho and H. Hoppe, “Poisson surface reconstruction,” in Proc. Eurographics Symposium on Geometry Processing, 2006. doi
  26. M. Kazhdan and H. Hoppe, “Screened Poisson surface reconstruction,” ACM Transactions on Graphics, vol. 32, no. 3, 2013. doi
  27. A. Guédon and V. Lepetit, “SuGaR: Surface-aligned Gaussian splatting for efficient 3D mesh reconstruction and high-quality mesh rendering,” in Proc. IEEE/CVF CVPR, 2024. CVF · arXiv
  28. B. Huang, Z. Yu, A. Chen, A. Geiger and S. Gao, “2D Gaussian splatting for geometrically accurate radiance fields,” in Proc. SIGGRAPH (Conference Papers), 2024. doi · arXiv
  29. S. Wang, V. Leroy, Y. Cabon, B. Chidlovskii and J. Revaud, “DUSt3R: Geometric 3D vision made easy,” in Proc. IEEE/CVF CVPR, 2024. CVF · arXiv
  30. V. Leroy, Y. Cabon and J. Revaud, “Grounding image matching in 3D with MASt3R,” in Proc. ECCV, 2024. doi · arXiv
  31. J. Wang, M. Chen, N. Karaev, A. Vedaldi, C. Rupprecht and D. Novotny, “VGGT: Visual geometry grounded transformer,” in Proc. IEEE/CVF CVPR, 2025. CVF · arXiv
  32. A. Pumarola, E. Corona, G. Pons-Moll and F. Moreno-Noguer, “D-NeRF: Neural radiance fields for dynamic scenes,” in Proc. IEEE/CVF CVPR, 2021. CVF
  33. G. Wu, T. Yi, J. Fang, L. Xie, X. Zhang, W. Wei, W. Liu, Q. Tian and X. Wang, “4D Gaussian splatting for real-time dynamic scene rendering,” in Proc. IEEE/CVF CVPR, 2024. CVF · arXiv
  34. Z. Wang, A. C. Bovik, H. R. Sheikh and E. P. Simoncelli, “Image quality assessment: From error visibility to structural similarity,” IEEE Transactions on Image Processing, vol. 13, no. 4, pp. 600–612, 2004. doi
  35. R. Zhang, P. Isola, A. A. Efros, E. Shechtman and O. Wang, “The unreasonable effectiveness of deep features as a perceptual metric,” in Proc. IEEE/CVF CVPR, 2018. CVF
  36. H. Fan, H. Su and L. J. Guibas, “A point set generation network for 3D object reconstruction from a single image,” in Proc. IEEE CVPR, 2017. CVF
  37. R. Jensen, A. Dahl, G. Vogiatzis, E. Tola and H. Aanæs, “Large scale multi-view stereopsis evaluation,” in Proc. IEEE CVPR, pp. 406–413, 2014. doi
  38. A. Knapitsch, J. Park, Q.-Y. Zhou and V. Koltun, “Tanks and Temples: Benchmarking large-scale scene reconstruction,” ACM Transactions on Graphics, vol. 36, no. 4, 2017. doi · benchmark
  39. J. T. Barron, B. Mildenhall, D. Verbin, P. P. Srinivasan and P. Hedman, “Mip-NeRF 360: Unbounded anti-aliased neural radiance fields,” in Proc. IEEE/CVF CVPR, 2022. CVF · arXiv
  40. D. Verbin, P. Hedman, B. Mildenhall, T. Zickler, J. T. Barron and P. P. Srinivasan, “Ref-NeRF: Structured view-dependent appearance for neural radiance fields,” in Proc. IEEE/CVF CVPR, 2022. CVF
  41. C.-H. Lin, W.-C. Ma, A. Torralba and S. Lucey, “BARF: Bundle-adjusting neural radiance fields,” in Proc. IEEE/CVF ICCV, 2021. CVF
  42. S. M. Seitz, B. Curless, J. Diebel, D. Scharstein and R. Szeliski, “A comparison and evaluation of multi-view stereo reconstruction algorithms,” in Proc. IEEE CVPR, vol. 1, pp. 519–528, 2006. doi
  43. Y. Furukawa and C. Hernández, “Multi-view stereo: A tutorial,” Foundations and Trends in Computer Graphics and Vision, vol. 9, no. 1–2, pp. 1–148, 2015. doi
  44. A. Tewari, J. Thies, B. Mildenhall, P. Srinivasan, E. Tretschk, W. Yifan, C. Lassner, V. Sitzmann et al., “Advances in neural rendering,” Computer Graphics Forum, vol. 41, no. 2, pp. 703–735, 2022. doi · arXiv
  45. G. Chen and W. Wang, “A survey on 3D Gaussian splatting,” ACM Computing Surveys, vol. 58, no. 12, 2026. doi · arXiv
  46. scikit-image contributors, “skimage.measure.marching_cubes,” scikit-image documentation (API reference). docs
  47. A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai et al., “An image is worth 16x16 words: Transformers for image recognition at scale,” in Proc. ICLR, 2021. arXiv
  48. A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser and I. Polosukhin, “Attention is all you need,” in Proc. NeurIPS, 2017. proceedings
  49. D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” in Proc. ICLR, 2015. arXiv