Orb_slam2: no camera data ingestion for realsense r200

Created on 4 Nov 2016  ·  10Comments  ·  Source: raulmur/ORB_SLAM2

Ubuntu 14.04 | 4.4.0-46-generic
ORB_SLAM2 1a3bc4bc9ca6501fd73042b0e05ae270ed7e6a0a

I've tried both monocular and rgbd with the same results. The app starts and I see a black screen with "waiting for images".

Screenshot:
http://imgur.com/a/1T01c

bash commands:

$ rosrun ORB_SLAM2 RGBD Vocabulary/ORBvoc.txt Examples/RGB-D/TUM1.yaml 

ORB-SLAM2 Copyright (C) 2014-2016 Raul Mur-Artal, University of Zaragoza.
This program comes with ABSOLUTELY NO WARRANTY;
This is free software, and you are welcome to redistribute it
under certain conditions. See LICENSE.txt.

Input sensor was set to: RGB-D

Loading ORB Vocabulary. This could take a while...
Vocabulary loaded!


Camera Parameters: 
- fx: 517.306
- fy: 516.469
- cx: 318.643
- cy: 255.314
- k1: 0.262383
- k2: -0.953104
- k3: 1.16331
- p1: -0.005358
- p2: 0.002628
- fps: 30
- color order: RGB (ignored if grayscale)

ORB Extractor Parameters: 
- Number of Features: 1000
- Scale Levels: 8
- Scale Factor: 1.2
- Initial Fast Threshold: 20
- Minimum Fast Threshold: 7

Depth Threshold (Close/Far Points): 3.09294

Most helpful comment

I tried orbslam2 with the kinect v1 and it worked fine so I decided it was a topic naming issue, which it was.

The depth topic orb slam 2 is looking for is /camera/depth_registered/image_raw and the freenect_launch specifies this whereas the r200_nodelet does not, it instead maps depth to /camera/depth/image_raw. You can change the launch files themselves if you want but the easiest way to get going is by running the following:

$ rosrun ORB_SLAM2 RGBD Vocabulary/ORBvoc.txt Examples/RGB-D/TUM1.yaml /camera/depth_registered/image_raw:=/camera/depth/image_raw

All 10 comments

You may have to provide a .bag file. See the explanations in Orb slam v.1

Στις 4 Νοε 2016 17:34, ο χρήστης "Jackson" [email protected]
έγραψε:

Ubuntu 14.04 | 4.4.0-46-generic
ORB_SLAM2 master

I've tried both monocular and rgbd with the same results. The app starts
and I see a black screen with "waiting for images".

Screenshot:
http://imgur.com/a/1T01c

bash commands:

$ rosrun ORB_SLAM2 RGBD Vocabulary/ORBvoc.txt Examples/RGB-D/TUM1.yaml

ORB-SLAM2 Copyright (C) 2014-2016 Raul Mur-Artal, University of Zaragoza.
This program comes with ABSOLUTELY NO WARRANTY;
This is free software, and you are welcome to redistribute it
under certain conditions. See LICENSE.txt.

Input sensor was set to: RGB-D

Loading ORB Vocabulary. This could take a while...
Vocabulary loaded!

Camera Parameters:

  • fx: 517.306
  • fy: 516.469
  • cx: 318.643
  • cy: 255.314
  • k1: 0.262383
  • k2: -0.953104
  • k3: 1.16331
  • p1: -0.005358
  • p2: 0.002628
  • fps: 30
  • color order: RGB (ignored if grayscale)

ORB Extractor Parameters:

  • Number of Features: 1000
  • Scale Levels: 8
  • Scale Factor: 1.2
  • Initial Fast Threshold: 20
  • Minimum Fast Threshold: 7

Depth Threshold (Close/Far Points): 3.09294


You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
https://github.com/raulmur/ORB_SLAM2/issues/190, or mute the thread
https://github.com/notifications/unsubscribe-auth/AVoPC17A7O4O0bYteegpXwqK4lM04_18ks5q61CSgaJpZM4KprFv
.

@geokarant A bag is a ros recording. Is orb slam 2 strictly offline only? I was under the impression that you could use it in realtime.

You are right, of course it s for real time but I haven't tried it yet! I
think there is a webcam ros package or something.. I f you figure it out
let me know, please! Thank you in advance!!

Στις 4 Νοε 2016 19:43, ο χρήστης "Jackson" [email protected]
έγραψε:

@geokarant https://github.com/geokarant A bag is a ros recording. Is
orb slam 2 strictly offline only? I was under the impression that you could
use it in realtime.


You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/raulmur/ORB_SLAM2/issues/190#issuecomment-258500164,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AVoPCz6cNdZF3EvzloCtKhtqMMaphWkpks5q627XgaJpZM4KprFv
.

I'm using an Intel RealSense r200. The package for that is ros-[distro]-realsense_camera. I'll give you an example with a kinect though:

sudo apt-get install ros-[distro]-openni-* # or freenect depending on which system you want
roscore & # you'll need ros installed, I recommend indigo http://wiki.ros.org/indigo/Installation
roslaunch openni_launch openni.launch # or freenect
rosrun ORB_SLAM2 RGBD Vocabulary/ORBvoc.txt Examples/RGB-D/TUM1.yaml

That SHOULD be all but something must be missing because you won't see the live camera feed, you'll see that the slam is "waiting for images"

Thank you very much, I ll try it too, and let you know!

Στις 4 Νοε 2016 20:08, ο χρήστης "Jackson" [email protected]
έγραψε:

I'm using an Intel RealSense r200. The package for that is
ros-[distro]-realsense_camera. I'll give you an example with a kinect
though:

sudo apt-get install ros-[distro]-openni-* # or freenect depending on which system you want
roscore & # you'll need ros installed, I recommend indigo http://wiki.ros.org/indigo/Installation
roslaunch openni_launch openni.launch # or freenect
rosrun ORB_SLAM2 RGBD Vocabulary/ORBvoc.txt Examples/RGB-D/TUM1.yaml

That SHOULD be all but something must be missing because you won't see the
live camera feed, you'll see that the slam is "waiting for images"


You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/raulmur/ORB_SLAM2/issues/190#issuecomment-258506606,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AVoPCwnGOEm0gytJ3osqXyMOzxhmbiHkks5q63SngaJpZM4KprFv
.

I tried orbslam2 with the kinect v1 and it worked fine so I decided it was a topic naming issue, which it was.

The depth topic orb slam 2 is looking for is /camera/depth_registered/image_raw and the freenect_launch specifies this whereas the r200_nodelet does not, it instead maps depth to /camera/depth/image_raw. You can change the launch files themselves if you want but the easiest way to get going is by running the following:

$ rosrun ORB_SLAM2 RGBD Vocabulary/ORBvoc.txt Examples/RGB-D/TUM1.yaml /camera/depth_registered/image_raw:=/camera/depth/image_raw

how to use usb camera for ros examples?

Search on ros wiki for webcam_node for ros. I havent tried it but i know it
exists

2016-11-25 14:10 GMT+02:00 Balaji-Govindaraj notifications@github.com:

how to use usb camera for ros examples?


You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/raulmur/ORB_SLAM2/issues/190#issuecomment-262943908,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AVoPC8_JkxOw8nhnEm_KNBRgR4fVmhemks5rBtBGgaJpZM4KprFv
.

I tried orbslam2 with the kinect v1 and it worked fine so I decided it was a topic naming issue, which it was.

The depth topic orb slam 2 is looking for is /camera/depth_registered/image_raw and the freenect_launch specifies this whereas the r200_nodelet does not, it instead maps depth to /camera/depth/image_raw. You can change the launch files themselves if you want but the easiest way to get going is by running the following:

$ rosrun ORB_SLAM2 RGBD Vocabulary/ORBvoc.txt Examples/RGB-D/TUM1.yaml /camera/depth_registered/image_raw:=/camera/depth/image_raw

HI @jacksonkr ,
I'm facing same problem as you were that orb-slam2 is running and showing blank screen "Waiting for images",
I don't know how to run it with real time, what commands and launch files i have to change. I'm using kinetic version of ROS, and my camera topics are /stereo/left/image_raw and /stereo/right/image_raw.

I have no idea of freenect_launch,
Please help me how can i run it with real time. I will appreciate your time.

@farhanmubasher This was a while ago so I can't quite remember the details. Either orb slam wouldn't work or I couldn't get it to do what I needed.

As far as freenect_launch is concerned it's part of the ros freenect package (ros-freenect / libfreenect) which can be used with the kinect v1.

Good luck!

Was this page helpful?
0 / 5 - 0 ratings

Related issues

zsy372901 picture zsy372901  ·  3Comments

akashshar picture akashshar  ·  4Comments

lokeshluckybisht picture lokeshluckybisht  ·  5Comments

Andreluizfc picture Andreluizfc  ·  4Comments

sunstarchan picture sunstarchan  ·  4Comments