You can visualize also with Rerun as an alternative or addition to normal plotting.

This can be for example useful to replay trajectories to visualize in more detail what is happening over time in an evaluation, plus the 3D visualization performance is 1000x times better in Rerun compared to Matplotlib.

You need to install the rerun-sdk in the Python environment where you have installed evo:

pip install rerun-sdk

Then you can use it by simply passing --rerun to your command or by using the evo.tools.rerun_bridge module in custom code (dummy example).

Here's an example of an absolute pose error evaluation with evo_ape using the EuRoC dataset trajectories from the test/data directory of this repository:

ℹ️ Transform axis markers share the same setting for the default axis length with the normal plots (see here). You can adapt the length also independently in the Rerun viewer by changing the axis_length for the transform entities in the selection panel.

⚠️ Rerun integration is still somewhat experimental. You will need at least evo v1.35.0 for good support.

See also the related issue.