Extracting correct data
Opened this issue · 3 comments
Hi,
I'm trying to use this camera to create a depth map. At first I thought I would need to do all the processing (getting the raw camera data, matching time stamps, finding correspondences) but I see that your package already sends out a lot of data to different topics. For the right camera:
/elp/right/camera_info
/elp/right/image_raw
/elp/right/image_raw/compressed
/elp/right/image_raw/compressed/parameter_descriptions
/elp/right/image_raw/compressed/parameter_updates
/elp/right/image_raw/compressedDepth
/elp/right/image_raw/compressedDepth/parameter_descriptions
/elp/right/image_raw/compressedDepth/parameter_updates
/elp/right/image_raw/theora
/elp/right/image_raw/theora/parameter_descriptions
/elp/right/image_raw/theora/parameter_updates
Could you briefly explain what these are and which one would be most helpful for me?
How does right / left camera each have a compressedDepth?
Somewhat related:
I looked into your code and it seems like "elp_stereo_camera.urdf.xacro" is the main program? But it seems so simple and I couldn't tell where/how you're publishing all these topics and how rviz is using the data.
Thank you for the help.
Hi,
So I can manage to read and display the "/elp/*/image_raw/compressed" but I cannot do the same with the "compressedDepth"
I get this warning when I try to echo "/elp/right/image_raw/compressedDepth"
Compressed Depth Image Transport - Compression requires single-channel 32bit-floating point or 16bit raw depth images (input format is: rgb8).
Waiting to see if you can explain what is being published to compressedDepth; or maybe I'm looking at the wrong topic?
Hey, I'm probably not going to get a chance to look up the topics until Monday. Feel free to ping me again then.
Josh
Sent from my iPhone
On Jul 21, 2016, at 4:18 PM, GnawXam notifications@github.com wrote:
Hi,
So I can manage to read and display the "/elp/*/image_raw/compressed" but I cannot do the same with the "compressedDepth"
I get this warning when I try to echo "/elp/right/image_raw/compressedDepth"
Compressed Depth Image Transport - Compression requires single-channel 32bit-floating point or 16bit raw depth images (input format is: rgb8).
Waiting to see if you can explain what is being published to compressedDepth; or maybe I'm looking at the wrong topic?
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.
Hi,
I never managed to use compressedDepth; I'm not sure what it does, actually. If you can clear some stuff up that'd be great.
Anyways, I'm using OpenCV's imshow() and I can display the CompressedImages in grayscale.
The goal is to get a depth map; I think I may need to know the focal length, sensor pixel size, and distance between the 2 cameras. Did you have a spec on these values? I think I found one on the distance between the 2 cameras but I don't know the other information.
Anything would help.
Thanks.