mlsanityData to 7 Oct 2026

Gradient-SDF: A Semi-Implicit Surface Representation for 3D Reconstruction

arXiv 2111.13652 ↗First posted 26 Nov 2021cs.CVcited by 12

Not on the map. Why?

Abstract (arXiv)

We present Gradient-SDF, a novel representation for 3D geometry that combines the advantages of implict and explicit representations. By storing at every voxel both the signed distance field as well as its gradient vector field, we enhance the capability of implicit representations with approaches originally formulated for explicit surfaces. As concrete examples, we show that (1) the Gradient-SDF allows us to perform direct SDF tracking from depth images, using efficient storage schemes like hash maps, and that (2) the Gradient-SDF representation enables us to perform photometric bundle adjustment directly in a voxel representation (without transforming into a point cloud or mesh), naturally a fully implicit optimization of geometry and camera poses and easy geometry upsampling. Experimental results confirm that this leads to significantly sharper reconstructions. Since the overall SDF voxel structure is still respected, the proposed Gradient-SDF is equally suited for (GPU) parallelization as related approaches.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 10 of its 35 references that are in the map. First posted on arXiv 26 Nov 2021.

Builds on

10 of 35 references.

References · 10

  1. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  2. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
  3. 2020Implicit Geometric Regularization for Learning Shapescited by 707
  4. 2021iMAP: Implicit Mapping and Positioning in Real-Timecited by 476
  5. 2021Neural RGB-D Surface Reconstructioncited by 312
  6. 2020Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfacescited by 146
  7. 2016A Large Dataset of Object Scanscited by 64
  8. 2020RoutedFusion: Learning Real-time Depth Map Fusioncited by 52
  9. 2020NeuralFusion: Online Depth Fusion in Latent Spacecited by 46
  10. 2018SurfelMeshing: Online Surfel-Based Mesh Reconstructioncited by 29

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 4; 2023: 3; 2024: 1; 2025: 3; 2026: 1.

12 papers.

  1. 2022Neural Vector Fields for Implicit Surface Representation and Inferencecited by 10
  2. 2022FIRe: Fast Inverse Rendering using Directional and Signed Distance Functionscited by 9
  3. 20254Deform: Neural Surface Deformation for Robust Shape Interpolationcited by 7
  4. 2025F-INR: Functional Tensor Decomposition for Implicit Neural Representationscited by 6
  5. 2022High-Quality RGB-D Reconstruction via Multi-View Uncalibrated Photometric Stereo and Gradient-SDFcited by 5
  6. 2022T4DT: Tensorizing Time for Learning Temporal 3D Visual Datacited by 2
  7. 2023Enhancing Surface Neural Implicits with Curvature-Guided Sampling and Uncertainty-Augmented Representationscited by 2
  8. 2023Cross-Dimensional Refined Learning for Real-Time 3D Visual Perception from Monocular Videocited by 1
  9. 2025OREN: Octree Residual Network for Real-Time Euclidean Signed Distance Mappingcited by 1
  10. 2023Coloring the Past: Neural Historical Buildings Reconstruction from Archival Photographycited by 0
  11. 2024TaylorGrid: Towards Fast and High-Quality Implicit Field Learning via Direct Taylor-based Grid Optimizationcited by 0
  12. 2026Hermite-NGP: Gradient-Augmented Hash Encoding for Learning PDEscited by 0