kitti2bag程式能夠講kitti原始資料(raw data)轉化為rosbag格式,以便于程式測試
一.安裝kitti2bag
步驟一:按照以下順序安裝kitti2bag程式
sudo pip install pandas==0.23.0 -i http://pypi.douban.com/simple/ --trusted-host pypi.douban.com
sudo pip install pykitti -i http://pypi.douban.com/simple --trusted-host pypi.douban.com
sudo pip install kitti2bag -i http://pypi.douban.com/simple --trusted-host pypi.douban.com
二.下載下傳KITTI原始資料集
前往如下網址下載下傳KITTI原始資料集(需要魔法):
http://www.cvlibs.net/datasets/kitti/raw_data.php
具體說明如下:
- 在頂端的Select category可以選擇場景,不用場景都有自己的資料集
- 每個資料集對應的Downloads由四個标簽,unsynced+unrectified date是沒有進行同步和矯正的資料
- synced+rectified是經過通過和矯正的資料
- calibration是标定資料,包含雷達,相機與imu之間的外參
- tracklets是軌迹
步驟二:選擇自己該興趣的包,下載下傳其synced+rectified檔案和calibration檔案
三.解壓資料集并調整檔案結構
步驟三:按照以下順序調整檔案結構
我們下載下傳下來的是兩個壓縮檔案
1.首先将其解壓
2.然後複制xxxx_xx_xx_calib/xxxx_xx_xx/下的三個标定檔案
3.将其粘貼到xxxx_xx_xx_drive_xxxx_sync/xxxx_xx_xx/檔案夾下,最終檔案結構如圖所示:
四.運作kitti2bag
步驟四:運作kitti2bag,講kitti資料集轉化為rosbag
1.打開終端,來到xxxx_xx_xx_drive_xxxx_sync/檔案夾下:
2.按照以下運作指令
kitti2bag -t 2011_09_26 -r 0027 raw_synced .
#2011_09_26是資料集對應的采集日期,在檔案夾名字中即可得到
#0027是資料集對應的編号,在synced+rectified檔案解壓後的檔案夾名稱中可以得到
#注意最後面有個空格和.
完整運作結果如圖:
[email protected]:~/Downloads/Downlods/2011_09_26_drive_0027_sync$ kitti2bag -t 2011_09_26 -r 0027 raw_synced .
Exporting static transformations
Exporting time dependent transformations
Exporting IMU
Exporting camera 0
100% (188 of 188) |#######################| Elapsed Time: 0:00:00 Time: 0:00:00
Exporting camera 1
100% (188 of 188) |#######################| Elapsed Time: 0:00:00 Time: 0:00:00
Exporting camera 2
100% (188 of 188) |#######################| Elapsed Time: 0:00:01 Time: 0:00:01
Exporting camera 3
100% (188 of 188) |#######################| Elapsed Time: 0:00:01 Time: 0:00:01
Exporting velodyne data
100% (188 of 188) |#######################| Elapsed Time: 0:00:24 Time: 0:00:24
## OVERVIEW ##
path: kitti_2011_09_26_drive_0027_synced.bag
version: 2.0
duration: 19.4s
start: Sep 26 2011 13:21:35.13 (1317014495.13)
end: Sep 26 2011 13:21:54.48 (1317014514.48)
size: 1020.9 MB
messages: 2632
compression: none [752/752 chunks]
types: geometry_msgs/TwistStamped [98d34b0043a2093cf9d9345ab6eef12e]
sensor_msgs/CameraInfo [c9a58c1b0b154e0e6da7578cb991d214]
sensor_msgs/Image [060021388200f6f0f447d0fcd9c64743]
sensor_msgs/Imu [6a62c6daae103f4ff57a132d6f95cec2]
sensor_msgs/NavSatFix [2d3a8cd499b9b4a0249fb98fd05cfa48]
sensor_msgs/PointCloud2 [1158d486dd51d683ce2f1be655c3c181]
tf2_msgs/TFMessage [94810edda583a504dfda3829e70d7eec]
topics: /kitti/camera_color_left/camera_info 188 msgs : sensor_msgs/CameraInfo
/kitti/camera_color_left/image_raw 188 msgs : sensor_msgs/Image
/kitti/camera_color_right/camera_info 188 msgs : sensor_msgs/CameraInfo
/kitti/camera_color_right/image_raw 188 msgs : sensor_msgs/Image
/kitti/camera_gray_left/camera_info 188 msgs : sensor_msgs/CameraInfo
/kitti/camera_gray_left/image_raw 188 msgs : sensor_msgs/Image
/kitti/camera_gray_right/camera_info 188 msgs : sensor_msgs/CameraInfo
/kitti/camera_gray_right/image_raw 188 msgs : sensor_msgs/Image
/kitti/oxts/gps/fix 188 msgs : sensor_msgs/NavSatFix
/kitti/oxts/gps/vel 188 msgs : geometry_msgs/TwistStamped
/kitti/oxts/imu 188 msgs : sensor_msgs/Imu
/kitti/velo/pointcloud 188 msgs : sensor_msgs/PointCloud2
/tf 188 msgs : tf2_msgs/TFMessage
/tf_static 188 msgs : tf2_msgs/TFMe
轉化成功後在xxxx_xx_xx_drive_xxxx_sync/檔案夾下生成對應的rosbag包
五.轉換結果測試
1.運作roscore
roscore
2.另開終端播放rosbag
rosbag play kitti_2011_09_26_drive_0027_synced.bag
3.列出topic清單可見下列topic
[email protected]:~$ rostopic list
/clock
/kitti/camera_color_left/camera_info
/kitti/camera_color_left/image_raw
/kitti/camera_color_right/camera_info
/kitti/camera_color_right/image_raw
/kitti/camera_gray_left/camera_info
/kitti/camera_gray_left/image_raw
/kitti/camera_gray_right/camera_info
/kitti/camera_gray_right/image_raw
/kitti/oxts/gps/fix
/kitti/oxts/gps/vel
/kitti/oxts/imu
/kitti/velo/pointcloud
/rosout
/rosout_agg
/tf
/tf_static
4.另開終端運作rviz
rviz
5.在rviz通過topic添加中可看到如下話題
6.添加各項話題,Fixed Frame調整為world,如下: