Visualize TF Frames
The MoveIt Pro 3D Visualizer can draw live TF coordinate frames as RGB axes, the same way RViz's TF display does. This is useful for inspecting the live pose of a frame — most commonly the robot's tip frame — without opening RViz, which is often impractical when working on a remote robot.
Frames are read from the /tf and /tf_static topics, so any frame your system publishes is available with no extra configuration.
Turn On TF Visualization
- In a 3D Visualizer pane, open the View menu in the top left.
- Under Display, enable TF Transforms.
Every published frame appears as a set of RGB axes — red for X, green for Y, blue for Z — drawn at the frame's live pose relative to the scene's reference frame.
Choose Which Frames to Show
A robot publishes many frames, so the 3D Visualizer sidebar lets you narrow the display. Open it from the settings icon on the TF Transforms row of the View menu, from All Display Settings at the bottom of that menu, from the cube icon next to the alerts bell, or by pressing V. Expand TF Transforms, then Frames:
- Frames lists every detected frame with a checkbox. Toggle individual frames on or off, or use the filter box to find a frame by name.
- Show All / Hide All toggle every frame at once — start from Hide All and enable just the frame you care about (for example, the tip frame).
Your selections are remembered across page reloads.
Adjust the Display
The settings nested under TF Transforms in the 3D Visualizer sidebar control how frames are drawn:
- Small Axes, Medium Axes, or Large Axes — pick the size that reads well for your robot's scale.
- Frame Labels — show or hide each frame's name next to its axes. Labels always face the camera and stay a fixed size on screen.
Read Labels Behind Geometry
With Frame Labels enabled, turn on Show TF Labels Through Geometry to read names hidden behind the robot or other scene geometry. The option is off by default and remembered in this browser across reloads. Turn it off to restore normal depth occlusion.
Only the text draws through geometry; the axes keep their normal depth occlusion. This changes visualization only, not TF poses or robot motion. Overlapping frame names can still cover one another; use the frame list to show fewer labels.
Frame poses are shown relative to the 3D Visualizer's reference frame. A frame that is not connected to the reference frame in the TF tree is hidden until a transform to it can be resolved.