ComfyUI-WanMove-Trajectory

ComfyUI ▸ Trajectory Drawer

Recreate motion trails in a single picture on top of any single image and export every intermediate frame.

https://github.com/user-attachments/assets/8e9dec19-28f2-4621-b072-f6c2db51eb29

https://github.com/user-attachments/assets/d738882e-668b-4f7f-a627-9952148d63b3

<img width="1046" height="834" alt="Image" src="https://github.com/user-attachments/assets/04e2fa15-4775-4c6d-9aa7-5481c272c3c6" /> <img width="932" height="1163" alt="Image" src="https://github.com/user-attachments/assets/4406aa93-d70a-441e-b134-02986c23c9bd" /> ## Highlights

• ✅ Shift + drag painting – The node ships with an interactive canvas: hold Shift, drag to paint, release to finish the stroke. No extra tools required.
• ✅ Customizable look – Tune stroke width, color, glow decay, sampling density, and background dim to match the neon trail aesthetics in the reference video.
• ✅ Per-frame output – The node returns an IMAGE batch representing the entire drawing animation, ready to be piped into SaveImage, VideoHelperSuite, etc.
• ✅ JSON trajectory – All stroke data is persisted as JSON inside the workflow so you can version, reuse, or share it effortlessly.

Installation

ComfyUI/custom_nodes/ComfyUI-Trajectory

Restart ComfyUI and the node will appear under “🧭 Trajectory”.

Node Parameters

| Parameter | Description | |-----------|-------------| | base_image | Image to draw on (only the first frame of the batch is used). | | path_json | Trajectory JSON produced by the built‑in editor. Click “Edit Trajectory (Shift Draw)” to open it. | | stroke_width | Stroke width in pixels. | | frames_per_segment | Extra interpolation samples inserted between adjacent points. Higher values = smoother trails. | | line_color | Stroke color (#00f0ff, r,g,b, etc.). | | tail_alpha | Opacity of the main stroke (0–1). | | background_dim | Darkens the base image before drawing so the trail pops out. | | trail_decay | Per-frame decay factor (default 0.3). Values <1 create persistent light trails; smaller values decay slower. | | line_style | Stroke style (solid or neon). | | fps | Output frame rate (use the same value when assembling a video). | | duration_seconds | Target animation duration. Set to 0 to derive from stroke length. | | show_pointer | Draw a pointer arrow in the output (for presentations/debug) without affecting the glow. |

Outputs:

• frames (IMAGE): the animation frames showing the trail being drawn.
• info (STRING): JSON metadata (fps, resolution, decay settings, etc.).
• tracks_json (STRING): the full list of pixel coordinates (e.g. [{"x":309,"y":338}, ...]) for downstream nodes or scripts.

Canvas Workflow

1. Click “Edit Trajectory (Shift Draw)” on the node.
2. Optionally load any PNG/JPG reference. The canvas automatically adopts the image’s original resolution and persists it for the next edit session.
3. Hold Shift and drag with the left mouse button to draw; release to finish a stroke.
4. Use Undo / Clear for quick fixes, Export PNG to snapshot the current view.
5. Hit Apply Path to store the JSON back into path_json, then run the workflow to render frames.

On touch devices you can draw directly. On desktop, holding Shift is required; otherwise the canvas shows a reminder.

Example Pipelines

1. LoadImage → TrajectoryDrawer → SaveImage (PNG sequence).
2. LoadImage → TrajectoryDrawer → VideoHelperSuite ▸ ImagesToVideo (render MP4/GIF).

Matching the Reference Video

• Import a still frame from large-motion-1-top.mp4 as the reference image to trace the exact path.
• Increase frames_per_segment, lower stroke_width, and set background_dim to ~0.2–0.4 to mimic the glowing trail look.

Limitations

• Only the first image in the batch is used. Split batches upstream if you need multiple trajectories.
• The JSON stores the reference image Base64 and size, but not the original file path—keep backups if necessary.

Feel free to extend the node (multi-color trails, speed curves, etc.) according to your workflow needs.