mlsanityData to 7 Oct 2026

MVPSNet: Fast Generalizable Multi-view Photometric Stereo

arXiv 2305.11167 ↗First posted 18 May 2023cs.CVcited by 18

Not on the map. Why?

Abstract (arXiv)

We propose a fast and generalizable solution to Multi-view Photometric Stereo (MVPS), called MVPSNet. The key to our approach is a feature extraction network that effectively combines images from the same view captured under multiple lighting conditions to extract geometric features from shading cues for stereo matching. We demonstrate these features, termed `Light Aggregated Feature Maps' (LAFM), are effective for feature matching even in textureless regions, where traditional multi-view stereo methods fail. Our method produces similar reconstruction results to PS-NeRF, a state-of-the-art MVPS method that optimizes a neural network per-scene, while being 411× faster (105 seconds vs. 12 hours) in inference. Additionally, we introduce a new synthetic dataset for MVPS, sMVPS, which is shown to be effective to train a generalizable MVPS method.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 33 of its 73 references that are in the map. First posted on arXiv 18 May 2023.

Builds on

33 of 73 references.

References · 33

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
  5. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,574
  6. 2020Accelerating 3D Deep Learning with PyTorch3Dcited by 781
  7. 2018MVSNet: Depth Inference for Unstructured Multi-view Stereocited by 780
  8. 2021UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstructioncited by 643
  9. 2020NeRV: Neural Reflectance and Visibility Fields for Relighting and View Synthesiscited by 493
  10. 2019Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matchingcited by 414
  11. 2021PhySG: Inverse Rendering with Spherical Gaussians for Physics-based Material Editing and Relightingcited by 332
  12. 2018DeepMVS: Learning Multi-view Stereopsiscited by 277
21 more
  1. 2019Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inferencecited by 252
  2. 2021NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Videocited by 241
  3. 2017SurfaceNet: An End-to-end 3D Neural Network for Multiview Stereopsiscited by 128
  4. 2021TransMVSNet: Global Context-aware Multi-view Stereo Network with Transformerscited by 117
  5. 2017SfSNet: Learning Shape, Reflectance and Illuminance of Faces in the Wildcited by 110
  6. 2022PS-NeRF: Neural Inverse Rendering for Multi-view Photometric Stereocited by 69
  7. 2019Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volumecited by 59
  8. 2020Two-shot Spatially-varying BRDF and Shape Estimationcited by 59
  9. 2020Multi-View Photometric Stereo: A Robust Solution and Benchmark Dataset for Spatially Varying Isotropic Materialscited by 56
  10. 2018PS-FCN: A Flexible Learning Framework for Photometric Stereocited by 51
  11. 2019Self-calibrating Deep Photometric Stereo Networkscited by 47
  12. 2021Shape and Material Capture at Homecited by 36
  13. 2022Uncertainty-Aware Deep Multi-View Photometric Stereocited by 30
  14. 2021Neural Radiance Fields Approach to Deep Multi-View Photometric Stereocited by 25
  15. 2017SilNet : Single- and Multi-View Reconstruction by Learning from Silhouettescited by 24
  16. 2020Deep Photometric Stereo for Non-Lambertian Surfacescited by 24
  17. 2021Curvature-guided dynamic scale networks for Multi-view Stereocited by 17
  18. 2022Multi-View Photometric Stereo Revisitedcited by 17
  19. 2022PS-Transformer: Learning Sparse Photometric Stereo Network using Self-Attention Mechanismcited by 13
  20. 2022Fast Light-Weight Near-Field Photometric Stereocited by 12
  21. 2022Neural apparent BRDF fields for multiview photometric stereocited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 4; 2024: 8; 2025: 4; 2026: 2.

18 papers.

Sort:

  1. 2023SuperNormal: Neural Surface Reconstruction via Multi-View Normal Integrationcited by 18
  2. 2024Towards 3D Vision with Low-Cost Single-Photon Camerascited by 15
  3. 2025MonoInstance: Enhancing Monocular Priors via Multi-view Instance Alignment for Neural Rendering and Reconstructioncited by 15
  4. 2024Leveraging Near-Field Lighting for Monocular Depth Estimation from Endoscopy Videoscited by 14
  5. 2023RNb-NeuS: Reflectance and Normal-based Multi-View 3D Reconstructioncited by 13
  6. 2024Enhancing Trust in LLMs: Algorithms for Comparing and Interpreting LLMscited by 9
  7. 2023Joint Depth Prediction and Semantic Segmentation with Multi-View SAMcited by 8
  8. 2023Differentiable Point-based Inverse Renderingcited by 7
  9. 2024Sparse Views, Near Light: A Practical Paradigm for Uncalibrated Point-light Photometric Stereocited by 5
  10. 2025Multi-view Surface Reconstruction Using Normal and Reflectance Cuescited by 5
  11. 2024NPLMV-PS: Neural Point-Light Multi-View Photometric Stereocited by 4
  12. 2024PRM: Photometric Stereo based Large Reconstruction Modelcited by 4
6 more
  1. 2026Fisheye3R: Adapting Unified 3D Feed-Forward Foundation Models to Fisheye Lensescited by 4
  2. 2024GaNI: Global and Near Field Illumination Aware Neural Inverse Renderingcited by 2
  3. 2024LUCES-MV: A Multi-View Dataset for Near-Field Point Light Source Photometric Stereocited by 2
  4. 2025Mobile Robotic Multi-View Photometric Stereocited by 1
  5. 2025GLOW: Global Illumination-Aware Inverse Rendering of Indoor Scenes Captured with Dynamic Co-Located Light & Cameracited by 0
  6. 2026EvalMVX: A Unified Benchmarking for Neural 3D Reconstruction under Diverse Multiview Setupscited by 0