Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

semantic image layers don't work with camera resolutions other than 640x320 #109

Closed
shin-ka opened this issue Jan 27, 2025 · 6 comments
Closed

Comments

@shin-ka
Copy link

shin-ka commented Jan 27, 2025

Hi,

In my project (in simulation) I am trying to generate semantic layers, similar to this launch file : turtlesim_semantic_image_example.launch. Basically the semantic node is segmenting a particular object in the scene and I am adding a semantic layer to the /elevation_map_raw topic which contains info about where this object is lying on the grid.

I am able to get correct correspondences between image and grid and the output semantic layer looks correct for image resolution 640x320. However, if I change the resolution to any other value, the object location shifts and eventually the semantic layer has no information of the object anymore.

I selected the resolution of 640x320 in my project based on the turtlebot example where the turtlebot has camera resolution of 640x320.

Could you please help me out what might be causing this?

@mktk1117
Copy link
Collaborator

mktk1117 commented Feb 5, 2025

Hi,

Did you change the corresponding camera info?

@shin-ka
Copy link
Author

shin-ka commented Feb 5, 2025

Hi, yes the camera info topic is always updated with every new camera resolution I test with.

@mktk1117
Copy link
Collaborator

mktk1117 commented Feb 5, 2025

Umm, that's a bit weird. Maybe it is not configured correctly to use that camera_info topic?
We have been using with different resolutions, and didn't have this issue.

@shin-ka
Copy link
Author

shin-ka commented Feb 5, 2025

When I try increasing the resolution, it appears as if the information in the semantic grid layer shifts towards the left and at 1280x720 resolution the grid is completely empty.
Looks like the projection of the image on the grid doesn't happen anymore.

Could you please point out which parts of the code I should look into to find out the error?

@mktk1117
Copy link
Collaborator

mktk1117 commented Feb 5, 2025

Could you check the Subscriber params?

For example these camera_info topic params.

You can do rosparam list | grep camera_info_topic_name
to listup the related parameters and check the values with rosparam get <param_name>

@shin-ka
Copy link
Author

shin-ka commented Feb 9, 2025

Hi, was able to solve the issue. There is a resize param that was set to 0.5 which was resizing the images. Changed it to 1.0 and it works with other resolutions.

Thanks for your support!

@shin-ka shin-ka closed this as completed Feb 9, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants