Adding a new odometer stream for python
shreyas2311 opened this issue · 9 comments
I would like to know how to add an odometer element similar to the acceleration and speed that appears on the circle example. I have tried adding metrics but they only seem render the data in a graphical form. Additionally, it seems like the existing odometer UI elements are mapped to /vehicle/acceleration and /vehicle/velocity streams by default. Is there any way to modify this. Also, is there any way to remove these elements if I wanted to?
Hi @shreyas2311 ,
Have you found a solution to your problem yet?
No I could not. Eventually we came up with our own UI and made the required change in the front end.
I see. I also need a very simple functionality like adding some arbitrary text on top of the objects (similar to object IDs) like this:
Do you have any idea on how this can be implemented in xviz? Thanks.
You can use the label primitive if you want to display texts
For example : https://avs.auto/playground/?sample_streams=text
I was already able to do it on the playground and now I want to do it in my local xviz installation. For that, I modified tracklets-converter.js file under /xviz/examples/converters/kitti/src/converters similar to the playground, but it doesn't seem to have any effect when I run the client. Do you have any guess on what might be the problem? Thank you.
Oh ok. I am unfamiliar with the JS version. I used the python-xviz-avs. In that, when you are building your UI streams, you would just have to add the same thing in the playground and it should work. However, I think it should be something similar for JS as well.
May I ask which script you use to run the server with python? I followed the readme file under python folder, but serve_scenario.py throws error:
Thank you!
Looks like the error you are having is on the import line itself ( import xviz_avs ) which seems like an issue with the path. Can you try having your script in the same folder as the xviz_avs library? This should give you an idea.
Oh, apparently it was not a path issue, but I had to install protobuf via pip install protobuf==3.20.*. This solved my problem. Thank you!