Closed jonathanslee4 closed 3 years ago
According to the SemanticKITTI website:
The label is a 32-bit unsigned integer (aka uint32_t) for each point, where the lower 16 bits correspond to the label.
In case you are using python, you could get some help for the script for visualizing the inference results. Specifically, in visualize.py a LaserScan is created after reading the values from the .label file for each scan. So something similar is needed for you.
In C++, there is also something similar in the rangenet library here.
Perfect answer, thanks.
If you still have problems, reopen the issue and let us know.
Awesome, thank you for the help! And I suppose this is more of a ROS/KITTI dataset question, but do either of you know how to convert from the PointCloud2 type into the .bin format of the KITTI dataset?
Does anyone know what the .label files (created after running infer) consist of? I can't open them using any sort of text editor but I need to know how they're structured so I can convert them into a valid ROS message type for a project I'm working on. Thank you!