Presentation

Presentation
PanoraMIS is a dataset of ultra-wide field of view images with accurate ground truth, created by the MIS laboratory of the Université de Picardie Jules Verne, France. The dataset covers a large range of panoramic cameras (catadioptric, twin-fisheye), robotic platforms (wheeled, aerial and industrial robots), and testing environments (indoors and outdoors). PanoraMIS is well suited to validate new image-based robot-motion estimation algorithms, including visual odometry, visual SLAM, and deep learning-based methods. For further information about the dataset, see the accompanying paper:
"PanoraMIS: An Ultra-wide Field of View Image Dataset for Vision-Based Robot-Motion Estimation", H.-E. Benseddik, F. Morbidi, G. Caron, The International Journal of Robotics Research, vol. 39, no. 9, pp. 1037-1051, August 2020 (pdf)

Downloads

Downloads

Info Sequences

about

Sequence 1

About sequence 1 :

  • Robot platform : Stäubli TX60
  • Rotation about an axis
  • Available data : poses.txt, times.txt
  • Size : 2.27 GB
  • Number of images : 3600
  • Calibration data : 6 images, CameraParams.txt


Sequence 2

About sequence 2 :
  • Robot platform : Pioneer 3-AT
  • Pure rotation
  • Available data : poses.txt, times.txt
  • Size : 66.3 MB
  • Number of images : 158
  • Calibration data : 8 images, CameraParams.txt

Sequence 3

About sequence 3 :
  • Robot platform : Stäubli TX60
  • Rotation about an axis
  • Available data : poses.txt
  • Size : 628 MB
  • Number of images : 720
  • Calibration data : 6 images, CameraParams.txt

Sequence 4

About sequence 4 :
  • Robot platform : Stäubli TX60
  • Rotation about three axes
  • Available data : poses.txt
  • Size : 83.7 MB
  • Number of images : 94
  • Calibration data : 6 images, CameraParams.txt

Sequence 5

About sequence 5 :
  • Robot platform : Pioneer 3-AT
  • Pure rotation
  • Available data : poses.txt, times.txt
  • Size : 79.3 MB
  • Number of images : 156
  • Calibration data : 8 images, CameraParams.txt

Sequence 6

About sequence 6 :
  • Robot platform : Pioneer 3-AT
  • Figure-of-eight trajectory
  • Available data : poses.txt, times.txt
  • Size : 155 MB
  • Number of images : 318
  • Calibration data : 8 images, CameraParams.txt

Sequence 7


About sequence 7 :
  • Robot platform : Seekur Jr
  • Available data : GPS.txt, IMUint.txt, IMUext.txt, poses.txt, times.txt
  • Size : 7.83 GB (L1), 4.31 GB (L2)
  • Number of images : 5442 (L1), 2578 (L2)
  • 3-D point clouds for L1 and L2
OpenVSLAM - L1
OpenVSLAM - L2

Sequence 8


About sequence 8 :
  • Robot platform : Parrot Disco
  • Available data : GPS.txt, IMUint.txt, times.txt
  • Size : 769 MB
  • MP4 video

Contact

contact
Laboratory

Robotic Perception group, MIS Laboratory, Université de Picardie Jules Verne

Location

14 Quai de la Somme, 80080 Amiens, France

E-mail

fabio.morbidi[at]u-picardie.fr

Phone

+33 (0)3 22 82 59 02

Reference

"PanoraMIS: An Ultra-wide Field of View Image Dataset for Vision-Based Robot-Motion Estimation",

H.-E. Benseddik, F. Morbidi, G. Caron, The International Journal of Robotics Research, vol. 39, no. 9, pp. 1037-1051, August 2020.

Funding

Funding
This work was carried out as part of the Interreg VA France (Channel) England ADAPT project Assistive Devices for empowering disAbled People through robotic Technologies. The Interreg FCE Programme is a European Territorial Cooperation programme that aims to fund high quality cooperation projects in the Channel border region between France and England. The Programme is funded by the European Regional Development Fund (ERDF).
This project was also funded in part by the Conseil Régional des Hauts-de-France and by the CPER (Contrat de Plan État-Région) 2015-2020.

Logo