Published April 25, 2024 | Version v3
Dataset Open

Data for article: Neural Graph Mapping for Dense SLAM with Efficient Loop Closure

Creators

  • 1. ROR icon KTH Royal Institute of Technology
  • 2. ROR icon Graz University of Technology

Description

Data to reproduce experiments in "Neural Graph Mapping for Dense SLAM with Efficient Loop Closure".

It contains:

  • a preprocessed version of the Kintinuous data from here (kintinuous_data.zip)
  • novel sequences on the larger Replica scenes (replica_big_data.zip)
  • ORB-SLAM2 results on
    • Kintinuous (kintinuous_slam.zip)
    • Replica Big (replica_big_slam.zip)
    • Replica iMAP (replica_slam.zip)
    • NRGBD (nrgbd_slam.zip)
    • ScanNet subset (scannet_subset_slam.zip)
  • Additional views for evaluation purposes as proposed by Co-SLAM for
    • Replica Big (replica_big_data.zip)
    • Replica iMAP (replica_coslam_eval.zip, modified from here with additional views for staircase and kitchen scenes, which were missing in the original data)
    • NRGBD (nrgbd_coslam_eval.zip, from here)

Files

kintinuous_data.zip

Files (10.2 GB)

Name Size Download all
md5:1ee73758267284db3cea47e3e16a5c62
2.2 GB Preview Download
md5:00dce104aecfbd67dfcf2013eb018aa6
289.7 MB Preview Download
md5:59aed99d910f2fd139c3029c331aeefc
45.0 MB Preview Download
md5:c8c6f8bfdcdd5c380f6ce807eeddc6c4
26.1 MB Preview Download
md5:d133cd91fb5323698af1c80d5712b3c8
7.3 GB Preview Download
md5:7abbc60bb4519c65c16fbd665e2259d4
51.5 MB Preview Download
md5:fd4b7f2a2a7b130dbfd187929fd186bf
170.8 MB Preview Download
md5:6998349547f460d2e8e86f13c4aca466
30.1 MB Preview Download
md5:4af0ead86873e1c508b4e40b6b3be0a5
36.3 MB Preview Download

Additional details

Software

Repository URL
https://github.com/KTH-RPL/neural_graph_mapping
Programming language
Python