Unable to get Organized Point Cloud , RGB Point Cloud from Isaac Sim via Omni.Ros2.Bridge in ROS2

HI,
I have a realsesne looking at my setup. Using the pseudo_depth input of the camera, I am able to get rgb, depth and the Point Cloud , and able to visualize them in Rviz2.
I am though unable to get RGB values, in my Point cloud, and also the Point Clouds are not oragnized. The is_dense, parameter in ros_msg is true.
Can someone please help me to get the Organized point cloud from isaac sim, and also is there a way to get the RGB value to from Isaac Sim directly.
I am using Action Graph for the logic.
My Isaac Sim version is - 2023.1.0.
Below is the Point Cloud Image that is generated.

ros-message - header:
stamp:
sec: 104
nanosec: 600005455
frame_id: sim_camera
height: 1
width: 921600
fields:

  • name: x
    offset: 0
    datatype: 7
    count: 1
  • name: y
    offset: 4
    datatype: 7
    count: 1
  • name: z
    offset: 8
    datatype: 7
    count: 1
    is_bigendian: false
    point_step: 12
    row_step: 11059200
    data:
  • 86
  • 38
  • 185
  • 192
  • 32
  • 75
  • 80
  • 192
  • 93
  • 112
  • 183
  • 64
  • 165
  • 139
  • 184
  • 192
  • 70
  • 240
  • 79
  • 192
  • 90
  • 32
  • 183
  • 64
  • 201
  • 241
  • 183
  • 192
  • 20
  • 150
  • 79
  • 192
  • 234
  • 208
  • 182
  • 64
  • 190
  • 88
  • 183
  • 192
  • 133
  • 60
  • 79
  • 192
  • 11
  • 130
  • 182
  • 64
  • 135
  • 192
  • 182
  • 192
  • 158
  • 227
  • 78
  • 192
  • 192
  • 51
  • 182
  • 64
  • 32
  • 41
  • 182
  • 192
  • 90
  • 139
  • 78
  • 192
  • 5
  • 230
  • 181
  • 64
  • 128
  • 146
  • 181
  • 192
  • 179
  • 51
  • 78
  • 192
  • 211
  • 152
  • 181
  • 64
  • 183
  • 252
  • 180
  • 192
  • 184
  • 220
  • 77
  • 192
  • 57
  • 76
  • 181
  • 64
  • 175
  • 103
  • 180
  • 192
  • 83
  • 134
  • 77
  • 192
  • 35
  • 0
  • 181
  • 64
  • 117
  • 211
  • 179
  • 192
  • 146
  • 48
  • 77
  • 192
  • 158
  • 180
  • 180
  • 64
  • 251
  • 63
  • 179
  • 192
  • 105
  • 219
  • 76
  • 192
  • ‘…’
    is_dense: true

Hi @rishavpathak23 I’m moving this question to the isaacsim forum, where I think you’ll be able to get answers specific to that app’s functionality.

Hi @rishavpathak23 can you please provide more details about what ‘Organized point cloud’ means?

For getting Images (RGB) you can use the ROS 2 camera helper node and set the topic type to RGB while publishing.

Hi @rchadha , In ROS message for realsense when you echo the topic , you will get a key “isdense”. If the following value is true, then it means it is un-orrganised point cloud meaning for a frame of 1024*720 not all points in the frame will be available. However if the following value if false, it means it contains all the pixel information for the above frame. This comes in handy when we need to do some model fitting for pick and place operations.

Hello, is their any update. I am also working with the simulated Intel Realsense but only get unorganized pcl via ROS. Is their a way to publish organized pcls?

Hi @d.kalteis,

Publishing organized pcl is not supported currently in the OmniGraph workflow.

However, in the standalone workflow, you could manually construct a pcl message and fill it in by grabbing the depth data from the sensor.

We will look into supporting organized point clouds from within the OmniGraph workflow in the future.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.