RGB2LIDAR: Towards Solving Large-Scale Cross-Modal Visual Localization

  title={RGB2LIDAR: Towards Solving Large-Scale Cross-Modal Visual Localization},
  author={Niluthpol Chowdhury Mithun and Karan Sikka and Han-Pang Chiu and S. Samarasekera and Rakesh Kumar},
  journal={Proceedings of the 28th ACM International Conference on Multimedia},
  • Niluthpol Chowdhury Mithun, Karan Sikka, +2 authors Rakesh Kumar
  • Published 2020
  • Computer Science
  • Proceedings of the 28th ACM International Conference on Multimedia
  • We study an important, yet largely unexplored problem of large-scale cross-modal visual localization by matching ground RGB images to a geo-referenced aerial LIDAR 3D point cloud (rendered as depth images). Prior works were demonstrated on small datasets and did not lend themselves to scaling up for large-scale applications. To enable large-scale evaluation, we introduce a new dataset containing over 550K pairs (covering 143 km2 area) of RGB and aerial LIDAR depth images. We propose a novel… CONTINUE READING

    Figures and Tables from this paper.


    Publications referenced by this paper.
    Wide-Area Image Geolocalization with Aerial Reference Imagery
    • 109
    • Highly Influential
    • PDF
    CVM-Net: Cross-View Matching Network for Image-Based Ground-to-Aerial Geo-Localization
    • 41
    • Highly Influential
    • PDF
    Image to LIDAR matching for geotagging in urban environments
    • 15
    • Highly Influential
    Semantic Visual Localization
    • 107
    • Highly Influential
    • PDF
    DublinCity: Annotated LiDAR Point Cloud and its Applications
    • 10
    • Highly Influential
    • PDF
    Geometric Urban Geo-localization
    • 32
    • Highly Influential
    • PDF