Privacy Preservance in Structure-from-Motion

The recent trend towards cloud-based localization and mapping systems has raised significant privacy concerns. These are mainly grounded by the fact that these services require users to upload visual data to their servers, which can reveal potentially confidential information, even if only derived image features are uploaded.

Our group has developped privacy-preserving techniques in aspects as Structure-from-Motion pipeline, map and query preservance in localization, privacy preserving for image features.

Image 1
Privacy Preserving Structure-from-Motion
Image 1
Privacy Preserving Querries


  • Privacy Preserving Image-Based Localization (CVPR 2019) [Paper]
  • Privacy Preserving Image Queries for Camera Localization (ICCV 2019) [Paper]
  • Privacy Preserving Structure-from-Motion (ECCV 2020) [Project page]
  • Privacy-Preserving Image Features via Adversarial Affine Subspace Embeddings (CVPR 2021) [Project page]
  • Privacy Preserving Localization and Mapping from Uncalibrated Cameras (CVPR 2021) [Project page]
  • Privacy Preserving Partial Localization (CVPR 2022) [Project page]
  • Privacy Preserving Localization via Coordinate Permutation (ICCV 2023) [Project page]