DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction

Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review


Deep Neural Networks (DNNs) have the potential to improve the quality of image-based 3D reconstructions. However, the use of DNNs in the context of 3D reconstruction from large and high-resolution image datasets is still an open challenge, due to memory and computational constraints. We propose a pipeline which takes advantage of DNNs to improve the quality of 3D reconstructions while being able to handle large and high-resolution datasets. In particular, we propose a confidence prediction network explicitly tailored for Multi-View Stereo (MVS) and we use it for both depth map outlier filtering and depth map refinement within our pipeline, in order to improve the quality of the final 3D reconstructions. We train our confidence prediction network on (semi-)dense ground truth depth maps from publicly available real world MVS datasets. With extensive experiments on popular benchmarks, we show that our overall pipeline can produce state-of-the-art 3D reconstructions, both qualitatively and quantitatively.

Original languageEnglish
Title of host publicationProceedings - 2020 International Conference on 3D Vision, 3DV 2020
Number of pages10
ISBN (Electronic)9781728181288
Publication statusPublished - 25 Nov 2020
Event8th International Conference on 3D Vision: 3DV 2020 - Online, Fukuoka, Virtual, Japan
Duration: 25 Nov 202028 Nov 2020

Publication series

NameProceedings - 2020 International Conference on 3D Vision, 3DV 2020


Conference8th International Conference on 3D Vision
Abbreviated title3DV
CityFukuoka, Virtual
Internet address


  • 3D Reconstruction
  • Confidence Prediction
  • Multi View Stereo

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition


Dive into the research topics of 'DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction'. Together they form a unique fingerprint.

Cite this