Göm meny

AMUSE

The automotive multi-sensor (AMUSE) dataset consists of inertial and other complementary sensor data combined with monocular, omnidirectional, high frame rate visual data taken in real traffic scenes during multiple test drives.

Quicklinks:

DOWNLOAD SAMPLE SEQUENCES

DATASET DISTRIBUTION

API DOWNLOAD

FURTHER INFORMATION

COPYRIGHT AND CONTACT

Scenes with different traffic and weather conditions as stitched and clipped panoramic images


DATASET DESCRIPTION

A short description of the dataset can be found in the paper published on CVPR 2013 workshop "Ground Truth - What is a good dataset?" . The paper can also be accessed here .

Please cite us, when using this dataset.

@INPROCEEDINGS{ amuse2013gt,
author = {Philipp Koschorrek and Tommaso Piccini and Per Öberg and
Michael Felsberg and Lars Nielsen and Rudolf Mester},
title = {A multi-sensor traffic scene dataset with omnidirectional video},
booktitle = {Ground Truth - What is a good dataset? CVPR Workshop 2013},
year = {2013}
}

DOWNLOAD SAMPLE SEQUENCES

The table below shows the actual list of recorded sequences. Sequences marked with ^ are not distributed via web. Only sample images are available. Further information about how to get these sequences can be found in the distribution section . Sequences marked with * can be downloaded fully. Those sequences are short ones for demonstration purpose. Therefore the driven way is minor and not mentioned.

Paper name Sequence
name
Length
[km]

#Frames
Size [GB]
Specifics
Seq1 20130227_CVL_0_
AroundLHouse^
0.7 4180 45 Loop closing, (nearly) static scene, partly snow
Seq2 20130227_CVL_1_
ToLambohov^
2.9 9826 106 real traffic, partly snow
Seq3
20130227_CVL_2_
RoundsInLambohov^
1.4 6689 72 Loop closing, suburban area, little traffic, low altitude of sun
Seq4
20130227_CVL_3_
GarnisonenToValla^
4.6 21912 236 Urban area, traffic, low altitude of sun
Seq5
20130319_CVL_0_
OnValla^
1.4 6469 70 Snow
Seq6 20130319_CVL_1_
VallaToTornby^
4.3 19844 210 Snow, urban area, traffic, water and snow on lens
Seq7 20130319_CVL_2_
TornbyToValla^
9.1 39830 430 Snow, urban area, traffic, water and snow on lens
- 20130530_CVL_0_
DrivingARound
*
- 720 8 Short sequence
-
20130530_CVL_1_
StraightForward
*
- 522 6 Short sequence
- 20130530_CVL_2_
Turn
*
- 674 7 Short sequence
- 20130530_CVL_3_
BackwardForward
*
-
631 7
Short sequence
- 20130530_CVL_4_
Braking
*
- 317 4
Short sequence
- 20130530_CVL_5_
Parking
*
-
680 7 Short sequence
- 20130530_CVL_6_
AroundRoundabout^
2.1 5146 55 Short sequence, real traffic
Was your download successful? Checksum file here


DATASET DISTRIBUTION

Due to the high frame rate and resolution of the cameras, the number of the cameras and the lossless raw image storage one whole sequence can easily reach several Gigabytes. Therefore a download of the sequences is hardly possible.

We offer instead to send the whole set of sequences on an external hard drive disk by a price of 4000 Swedish Krona(SEK) incl. worldwide shipping (as of 2013-06-23, this price may change). Feel free to contact us .

Note: Shipping starts as soon as possible. Information about shipping can be found here.


API DOWNLOAD

For reading the sensor data stored in streamfiles an API is needed, the StreamReader . There are different APIs for different programming languages. The specific APIs can downloaded by clicking on the corresponding programming language. Documentation of the functions is inline. A common description of the functions can be found in the C/C++-documentation .

C/C++

Python

Matlab

ROS

For initialization the API needs the paths to the project on the one hand and to a XML file on the other hand which describes the IDs and messages used in the streamfiles. These paths can be committed directly or stored in a XML file which is used instead (more information in the API documentation). Below are the mandatory message/ID XML file and an example file of the initialization XML file.

Init XML file

Message/ID XML file



FURTHER INFORMATION

Furthermore, information about the experimental setup, the stream format, etc. can be found in the links below.

Sketches of the experimental setup

Detailed sensor description incl. camera calibration parameter

Stream format description


GUIDELINES FOR EXPANDING OUR WORK

Information to be published soon.


COPYRIGHT AND CONTACT


Creative Commons License
To get in touch with the people behind AMUSE use the mail address

amuse.dataset@isy.liu.se


Senast uppdaterad: 2014-05-12