NavVis Indoor Dataset

NavVis Indoor Dataset Overview

Basic Information

• Name: NavVis Indoor Dataset
• Description: Contains over 50,000 high‑resolution geo‑referenced images, covering more than 50,000 m² of indoor space across twelve buildings at the Technical University of Munich.
• Time Span: Recorded from August 2015 to March 2016.
• Features: Includes a variety of indoor spaces, different building styles, and lighting conditions.

TUM LSI Dataset

• Description: TUM LSI is a subset of the NavVis Indoor Dataset, containing 1,314 high‑resolution images covering a full floor of an office building at the Technical University of Munich, spanning 5,575 m².
• Scan ID: 2015-08-16_15.34.11
• Usage: Walch et al. used only the cam0‑cam4 cameras, resulting in 1,095 images for evaluation.

Organization

• Structure: Data is organized by consecutive scans.
• Storage: Images and corresponding poses are stored in images and poses directories, respectively.
• Identification: Each scan is identified by its unique timestamp <scan_timestamp>.
• Image Grouping: Images are grouped in sets of six taken at the same time and location, numbered from cam0 to cam5.

Data Format

• Images:
  • Format: JPEG
  • Size: 3448 × 4592 pixels
  • DPI: 180 dpi
  • Color Space: sRGB
• Poses:
  • Format: XML
  • Global Georeference: WGS84
  • Scan Coordinate System: 6‑DoF transform relative to the root node
  • Image Pose: 6‑DoF transform relative to the scan coordinate system

Access

• Contact: Data can be obtained by emailing research@navvis.com.