Missing support on sensor_msgs::image_encodings::TYPE_16UC1

Hi

Missing support on sensor_msgs::image_encodings::TYPE_16UC1, as realsense ros is outputting 16UC1 format depth image. Will there be release that support this image encodings ?

Below error is thrown when using nvidia::isaac_ros::depth_image_proc::PointCloudXyzNode and it’s subscribing to realsense camera topic /camera/depth/image_rect_raw .
However, i tried with using nvidia::isaac_ros::image_proc::ImageFormatConverterNode to convert /camera/depth/image_rect_raw to type nv12 , also suffer the same error.

[component_container_mt-3] [ERROR] [1724137462.210540163] [NitrosImage]: [convert_to_custom] Unsupported encoding from ROS [16UC1].
[component_container_mt-3] terminate called after throwing an instance of 'std::runtime_error'
[component_container_mt-3]   what():  [convert_to_custom] Unsupported encoding from ROS

No idea if isaac_ros_nitros or isaac_ros_depth_image_proc need support added for this TYPE_16UC1.

Best,
Samuel

Hi @HappySamuel

Thank you for your message. This format is not supported in Isaac ROS and cannot be accelerated on Nitros. This is likely the reason for the error you’re experiencing.

I’ve raised this issue with the engineering team to make a note of it for future implementation.

Why are you working with this format with your realsense camera? If you are working with other Isaac ROS packages, you can use other formats.

Let me know.
Raffaello

Hi @Raffaello

I checked with realsense-ros , they mentioned that realsense camera depth image only support Z16 (rs2 format) and output msg is Type_16UC1
https://github.com/IntelRealSense/realsense-ros/issues/3188#issuecomment-2298418205

The reason isaac_ros_depth_image_proc was chosen is because of using cuvslam, hence i have to turn OFF emitter, which is originally required for generating pointcloud. Therefore, i plan to use isaac_ros_depth_image_proc to convert the depth image to pointcloud. But faced above issue.

Best,
Samuel

Thank you @HappySamuel

About the Realsense, we also listed a set of errors that can be fixed. You can read the troubleshooting page:

https://nvidia-isaac-ros.github.io/troubleshooting/hardware_setup.html