mlsanityData to 7 Oct 2026

DI-Fusion: Online Implicit 3D Reconstruction with Deep Priors

arXiv 2012.05551 ↗First posted 10 Dec 2020cs.CV cs.GR cs.ROcited by 70

Not on the map. Why?

Abstract (arXiv)

Previous online 3D dense reconstruction methods struggle to achieve the balance between memory storage and surface quality, largely due to the usage of stagnant underlying geometry representation, such as TSDF (truncated signed distance functions) or surfels, without any knowledge of the scene priors. In this paper, we present DI-Fusion (Deep Implicit Fusion), based on a novel 3D representation, i.e. Probabilistic Local Implicit Voxels (PLIVoxs), for online 3D reconstruction with a commodity RGB-D camera. Our PLIVox encodes scene priors considering both the local geometry and uncertainty parameterized by a deep neural network. With such deep priors, we are able to perform online implicit 3D reconstruction achieving state-of-the-art camera trajectory estimation accuracy and mapping quality, while achieving better storage efficiency compared with previous online 3D reconstruction approaches. Our implementation is available at https://www.github.com/huangjh-pub/di-fusion.

Key facts. Cited by 70 papers in the map (data through 7 Oct 2026). First posted on arXiv 10 Dec 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 6; 2022: 18; 2023: 21; 2024: 14; 2025: 6; 2026: 5.

70 papers.

Sort:

  1. 2021NICE-SLAM: Neural Implicit Scalable Encoding for SLAMcited by 586
  2. 2022MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstructioncited by 467
  3. 2023Gaussian Splatting SLAMcited by 410
  4. 2023GS-SLAM: Dense Visual SLAM with 3D Gaussian Splattingcited by 278
  5. 2023Point-SLAM: Dense Neural Point Cloud-based SLAMcited by 160
  6. 2023Gaussian-SLAM: Photo-realistic Dense SLAM with Gaussian Splattingcited by 152
  7. 2022Vox-Fusion: Dense Tracking and Mapping with Voxel-based Neural Implicit Representationcited by 149
  8. 2023GO-SLAM: Global Optimization for Consistent 3D Instant Reconstructioncited by 132
  9. 20214DComplete: Non-Rigid Motion Estimation Beyond the Observable Surfacecited by 123
  10. 2023Neural Kernel Surface Reconstructioncited by 102
  11. 2024SLAM3R: Real-Time Dense Scene Reconstruction from Monocular RGB Videoscited by 100
  12. 2022Orbeez-SLAM: A Real-time Monocular Visual SLAM with ORB Features and NeRF-realized Mappingcited by 79
38 more
  1. 2022Towards Implicit Text-Guided 3D Shape Generationcited by 68
  2. 2022Physical Adversarial Attack meets Computer Vision: A Decade Surveycited by 66
  3. 2024LoopSplat: Loop Closure by Registering 3D Gaussian Splatscited by 59
  4. 2024Loopy-SLAM: Dense Neural SLAM with Loop Closurescited by 48
  5. 2022SHINE-Mapping: Large-Scale 3D Mapping Using Sparse Hierarchical Implicit Neural Representationscited by 44
  6. 2023Dense RGB SLAM with Neural Implicit Mapscited by 42
  7. 2024PIN-SLAM: LiDAR SLAM Using a Point-Based Implicit Neural Representation for Achieving Global Map Consistencycited by 42
  8. 2022BNV-Fusion: Dense 3D Reconstruction using Bi-level Neural Volume Fusioncited by 29
  9. 2024NeSLAM: Neural Implicit Mapping and Self-Supervised Feature Tracking With Depth Completion and Denoisingcited by 29
  10. 2023Neural Implicit Dense Semantic SLAMcited by 26
  11. 2021Correspondence-Free Point Cloud Registration with SO(3)-Equivariant Implicit Shape Representationscited by 24
  12. 2023MIPS-Fusion: Multi-Implicit-Submaps for Scalable and Robust Online Neural RGB-D Reconstructioncited by 24
  13. 2023DNS SLAM: Dense Neural Semantic-Informed SLAMcited by 23
  14. 2024DDN-SLAM: Real-time Dense Dynamic Neural Implicit SLAMcited by 22
  15. 2022PINs: Progressive Implicit Networks for Multi-Scale Neural Representationscited by 19
  16. 2023UncLe-SLAM: Uncertainty Learning for Dense Neural SLAMcited by 19
  17. 2024NIS-SLAM: Neural Implicit Semantic RGB-D SLAM for 3D Consistent Scene Understandingcited by 19
  18. 2022MonoNeuralFusion: Online Monocular Neural 3D Reconstruction with Geometric Priorscited by 17
  19. 2022ObjectMatch: Robust Registration using Canonical Object Correspondencescited by 15
  20. 2021NeuralBlox: Real-Time Neural Representation Fusion for Robust Volumetric Mappingcited by 14
  21. 2022Multi-sensor large-scale dataset for multi-view 3D reconstructioncited by 14
  22. 2025DenseSplat: Densifying Gaussian Splatting SLAM with Neural Radiance Priorcited by 14
  23. 2023Learning Neural Implicit through Volume Rendering with Attentive Depth Fusion Priorscited by 11
  24. 2025MCN-SLAM: Multi-Agent Collaborative Neural SLAM with Hybrid Implicit Neural Scene Representationcited by 11
  25. 2024NID-SLAM: Neural Implicit Representation-based RGB-D SLAM in dynamic environmentscited by 10
  26. 2024RTG-SLAM: Real-time 3D Reconstruction at Scale using Gaussian Splattingcited by 10
  27. 2022What's the Situation with Intelligent Mesh Generation: A Survey and Perspectivescited by 9
  28. 2022HRBF-Fusion: Accurate 3D reconstruction from RGB-D data using on-the-fly implicitscited by 8
  29. 2023SC-NeuS: Consistent Neural Surface Reconstruction from Sparse and Noisy Viewscited by 8
  30. 2022Robust Change Detection Based on Neural Descriptor Fieldscited by 7
  31. 2024Vox-Fusion++: Voxel-based Neural Implicit Dense Tracking and Mapping with Multi-mapscited by 7
  32. 2022Learning Online Multi-Sensor Depth Fusioncited by 6
  33. 2022SC-Explorer: Incremental 3D Scene Completion for Safe and Efficient Exploration Mapping and Planningcited by 6
  34. 2022An Algorithm for the SE(3)-Transformation on Neural Implicit Maps for Remapping Functionscited by 5
  35. 2023Uni-Fusion: Universal Continuous Mappingcited by 5
  36. 2025GauS-SLAM: Dense RGB-D SLAM with Gaussian Surfelscited by 5
  37. 2021CIRCLE: Convolutional Implicit Reconstruction and Completion for Large-scale Indoor Scenecited by 4
  38. 20233D Reconstruction with Generalizable Neural Fields using Scene Priorscited by 4

50 most cited of 70 All 70 (JSON)