Nodes/ComfyUI/WanMoveVisualizeTracks
ComfyUI Node Runs on cloud

WanMoveVisualizeTracks

See your motion tracks before you waste a generation on them

By Comfy-Org·Created 4 years ago·Updated a day ago· 130,493
WanMoveVisualizeTracks
  • images
  • tracks
  • IMAGE
line_resolution24
circle_size12
opacity0.75
line_width16

TRACKS data is invisible until you render it. WanMoveVisualizeTracks is the renderer: it takes your frames plus a track set and draws the trajectories on top - colored circles at each tracked point, with gradient trails showing where each point has been. You see exactly what WanMoveTrackToVideo is going to be conditioned on, before you commit a multi-minute generation to it. In a workflow where tracks are the whole point, this is the node you reach for first and delete last.

How it works

For each frame it draws a filled circle at every track point (each track gets a color, cycling through five) and a polyline tracing the track's history. The gradient on the trail fades from the current position backward, and line_resolution sets how many frames of history each trail covers. Everything lands on an RGBA overlay blended at opacity, so the original video stays readable underneath. Output is an IMAGE - a video you can preview or save.

The inputs

  • images - the frames to draw on. Usually your tiled start image (repeat it to clip length) or an already-generated video you want to audit.
  • tracks - optional. Disconnect it and the node just returns images unchanged.
  • line_resolution - frames of history drawn per trail (default 24).
  • circle_size - radius of the point markers (default 12).
  • opacity - 0 to 1, how strongly the overlay blends over the image.
  • line_width - thickness of the trails (default 16).

Output: IMAGE - the annotated video.

Getting it

Ships with ComfyUI core; it arrived with the Wan-Move family in December 2025. Nothing to install.

Practical notes

If your images have fewer frames than the tracks span, the node repeats the image frames to match - so you can preview a full-length track set against a single still. The five-color cycle means track #6 shares color with track #1; fine for debugging, mildly annoying if you're doing serious multi-object choreography, and the reason you'll want WanMoveConcatTrack plus careful ordering to keep related tracks on distinct colors. And it's a debug tool, not a preview of output: the tracks it shows are the input control, drawn on top of your source frames. It cannot tell you what the model will do with them - only what you asked for.

Categorymodel/conditioning/wan/move

Inputs (6)

NameTypeDefaultDescription
imagesIMAGE
line_resolutionINT241–1024
circle_sizeINT121–128
opacityFLOAT0.750–1
line_widthINT161–128
tracksoptTRACKS

Outputs (1)

NameTypeDescription
IMAGEIMAGE