Hi @ykim,
the root cause is that the image_stream script explicitly turns on the front light. This way the FrontLight parameter in the JSON file is always overwritten. I pushed a commit with which you should be able to disable the front light during capturing the texture image. Add "front_light": False, to the parameters of the image_stream Node in your launch file (ensenso.launch.py).
You can of course also write your own script if you like.
About TextureCapture:
The ensenso_driver does not compute the texture with cmdComputeTexture as it is done by NxView or Engage. The texture is “computed” exactly as you have done it in your launch file with a combination of the image_stream script and the texture_point_cloud Node (as also shown in our texture_point_cloud.launch file).
In your other post you said:
Also, I need to capture the texture image, so disabling Capture Texture won’t fit my use case.
As you are working with ROS, you can neglect the Capture/Texture parameters in the JSON and disable it, because you are capturing the texture images via the image_stream script, which performs the RequestData action the second time after your request_data_node.
By the way: the request_data script has no request_point_cloud parameter, only a point_cloud parameter. The parameter that you set in your launch file is simply ignored, but the point cloud is still requested by default.
Let me know if you have further questions.
Kind regards
Benny