Accelerated Cross-platform Implementation of Differentiable Rendering of Signed Distance Function

Reconstructing scene parameters from images (inverse rendering) is a popular problem and an important area in computer graphics and vision. Differentiable rendering, based on gradient optimization methods, is currently being increasingly applied to this task. This paper presents improvements to the method for differentiable rendering of signed distance functions, proposed in 2024, as well as a cross-platform implementation that supports execution on various types of graphics accelerators. This ensures independence from specific hardware vendors and expands the applicability of the method to heterogeneous hardware configurations. Our paper proposes two key modifications. First, we replace the standard ray tracing method with Newton's method and an analytical method adapted to differentiable rendering problems. Furthermore, we split the calculation of derivatives with respect to texture and geometric scene parameters into two parts, corresponding to the interior and boundary integrals. This partitioning reduces the number of Monte Carlo samples required to estimate texture gradients and allows the computation to be distributed between two shaders. As a result, the developed implementation of differentiable rendering is three times faster compared to the baseline implementation while maintaining the same level of accuracy.

1. Loper M.M., Black M.J. OpenDR: An Approximate Differentiable Renderer, in Computer Vision. ECCV 2014, Zurich, Switzerland, 2014.

2. 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, 2021, vol. 65, no. 1, pp. 99–106.

3. Kerbl B., Kopanas G., Leimkuehler T., Drettakis G. 3D Gaussian Splatting for Real-Time Radiance Field Rendering. ACM Trans. Graph., 2023, vol. 42, no. 4, pp. 1-14.

4. 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.

5. Müller T., Evans A., Schied C., Keller A. Instant neural graphics primitives with a multiresolution hash encoding. ACM Trans. Graph., 2022, vol. 41, no. 4, pp. 1-15.

6. Li Z., Müller T., Evans A., Taylor R. H., Unberath M., Liu M.-Y., Lin C.-H. Neuralangelo: High-Fidelity Neural Surface Reconstruction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, 2023.

7. Park J.J., Florence P., Straub J., Newcombe R., Lovegrove S. DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, California, 2019.

8. Wang Y., Han Q., Habermann M., Daniilidis K., Theobalt C., Liu L. NeuS2: Fast Learning of Neural Implicit Surfaces for Multi-view Reconstruction. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Paris, 2023.

9. Dogaru A., Ardelean A.T., Ignatyev S., Zakharov E., Burnaev E. Sphere-Guided Training of Neural Implicit Surfaces. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Los Alamitos, California, 2023.

10. Li T.-M., Aittala M., Durand F., Lehtinen J. Differentiable Monte Carlo ray tracing through edge sampling. ACM Trans. Graph., 2018, vol. 37, no. 6, pp. 1-11.

11. Loubet G., Holzschuch N., Jakob W. Reparameterizing discontinuous integrands for differentiable rendering. ACM Trans. Graph., 2019, vol. 38, no. 6, pp. 1-14.

12. Bangaru S.P., Li T.-M., Durand F. Unbiased warped-area sampling for differentiable rendering. ACM Trans. Graph., 2020, vol. 39, no. 6, pp. 1-18.

13. Nicolet B., Jacobson A., Jakob W. Large steps in inverse rendering of geometry. ACM Trans. Graph, 2021., vol. 40, no. 6, pp. 1-13.

14. Vicini D., Speierer S., Jakob W. Differentiable signed distance function rendering. ACM Trans. Graph., 2022, vol. 41, no. 4, p. 18.

15. Sethian J.A. A Fast Marching Level Set Method for Monotonically Advancing Fronts. In National Academy of Sciences of the United States of America, Washington, 1996.

16. Zhao H. A fast sweeping method for Eikonal equations. Math. Comp., 2005, vol. 74, pp. 603-627.

17. Wang Z., Deng X., Zhang Z., Jakob W., Marschner S. A Simple Approach to Differentiable Rendering of SDFs. In SIGGRAPH Asia 2024 Conference Papers, Tokyo, 2024.

18. Wang Z. The official implementation for 'A Simple Approach to Differentiable Rendering of SDFs'. Available at: https://github.com/zichenwang01/relaxed-boundary, accessed: 03. 02. 2026.

19. Luan F., Zhao S., Bala K., Dong Z. Unified Shape and SVBRDF Recovery using Differentiable Monte Carlo Rendering. Computer Graphics Forum, 2021, vol. 40, no. 4, pp. 101-113.

20. Hart J. Sphere Tracing: A Geometric Method for the Antialiased Ray Tracing of Implicit Surfaces. The Visual Computer, 1996, vol. 12, pp. 527–545.

21. Söderlund H.H., Evans A., Akenine-Möller T. Ray Tracing of Signed Distance Function Grids. Journal of Computer Graphics Techniques (JCGT), 2022, vol. 11, no. 3, pp. 94-113.

22. Budak A.S., Garifullin A.R., Frolov V.A., Galaktionov V.A. Study of Surface Representation Methods Based on Signed Distance Functions. Programming and Computer Software, 2025, vol. 51, pp. 131-139.

23. Frolov V.A., Sanzharov V.V., Galaktionov V.A. Kernel_slicer: high-level approach on top of GPU programming API. In Ivannikov Ispras Open Conference (ISPRAS OPEN), Moscow, 2022.

24. Detrixhe M., Gibou F., Min C. A parallel fast sweeping method for the Eikonal equation. Journal of Computational Physics, 2013, vol. 237, pp. 46-55.

25. Stanford Computer Graphics Laboratory. The Stanford 3D Scanning Repository. Available at: https://graphics.stanford.edu/data/3Dscanrep/, accessed: 03.02.2026.