Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

TerraSky3D: Multi-View Reconstructions of European Landmarks in 4K

Mattia D'Urso, Yuxi Hu, Christian Sormann, Mattia Rossi, Friedrich Fraundorfer

Abstract

Despite the growing need for data of more and more sophisticated 3D reconstruction pipelines, we can still observe a scarcity of suitable public datasets. Existing 3D datasets are either low resolution, limited to a small amount of scenes, based on images of varying quality because retrieved from the internet, or limited to specific capturing scenarios. Motivated by this lack of suitable 3D datasets, we captured TerraSky3D, a high-resolution large-scale 3D reconstruction dataset comprising 50,000 images divided into 150 ground, aerial, and mixed scenes. The dataset focuses on European landmarks and comes with curated calibration data, camera poses, and depth maps. TerraSky3D tries to answer the need for challenging dataset that can be used to train and evaluate 3D reconstruction-related pipelines.

TerraSky3D: Multi-View Reconstructions of European Landmarks in 4K

Abstract

Despite the growing need for data of more and more sophisticated 3D reconstruction pipelines, we can still observe a scarcity of suitable public datasets. Existing 3D datasets are either low resolution, limited to a small amount of scenes, based on images of varying quality because retrieved from the internet, or limited to specific capturing scenarios. Motivated by this lack of suitable 3D datasets, we captured TerraSky3D, a high-resolution large-scale 3D reconstruction dataset comprising 50,000 images divided into 150 ground, aerial, and mixed scenes. The dataset focuses on European landmarks and comes with curated calibration data, camera poses, and depth maps. TerraSky3D tries to answer the need for challenging dataset that can be used to train and evaluate 3D reconstruction-related pipelines.

Paper Structure

This paper contains 18 sections, 2 equations, 7 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Large-Scale Outdoor Photo Collections.
  4. Mixed Aerial and Ground data.
  5. Stereo Pose Benchmarks.
  6. Pipeline
  7. Data Collection and Scene Diversity
  8. Geometry Estimation
  9. Depth Computation
  10. Format and Split
  11. Experiments
  12. Stereo Pose Estimation
  13. End-to-End 3D Reconstruction
  14. Novel View Synthesis
  15. Bidirectional Geometric Consistency
  16. ...and 3 more sections

Figures (7)

  • Figure 1: Example scene from TerraSky3D. Left: Sparse reconstruction of the Villalta Castle, Italy. Right: Representative images collected from aerial and ground perspectives, shown with their corresponding semantically filtered depth maps.
  • Figure 2: Geographical Distribution of Data Collection Sites in TerraSky3D. The dataset includes sites across 11 European countries. Close locations might share the same red pin.
  • Figure 3: Example Scene from TerraSky3D. Left: Sparse reconstruction of the Natural History Museum, Vienna, Austria. Right: The first row shows example images, and the second row shows the corresponding semantically filtered depth maps.
  • Figure 4: Visualization of the Depth Filtering Process. From left to right: original RGB image, raw depth map from APD-MVS, semantic mask, APD-MVS confidence mask, and the filtered depth map. The scene depicted is the Arch of Victory, Madrid, Spain.
  • Figure 5: Example Scene from TerraSky3D. Left: Sparse reconstruction of the Italian Charnel House, Kobarid, Slovenia. Right: Representative images collected from aerial and ground perspectives, shown with their corresponding semantically filtered depth maps.
  • ...and 2 more figures