Skip to main content
Video Resize

Video Resize · screenshot coming soon

​
What this node does

Video Resize scales the full clip to width × height. Set width or height to -1 to derive from the other dimension and source aspect ratio (ffmpeg-style). Use it to normalize resolution or shrink file size. It does not run text/image-to-video workflows. PyAV backend; no GPU.

​
When to use it

  • Scale generations to 720p / 1080p for delivery
  • Standardize size after Crop
  • Smaller files before Demux / separation (Demucs pending)

​
How it works

  • Stage + ▶ Run; upstream snapshot.
  • PyAV resize_video; fps preserved.
  • At least one dimension must be positive; both -1 is invalid.

​
Types

bridges.md

​
Parameters

​
video

Source COMFYTV_VIDEO.

​
width / height

Target pixels, default 1280×720. Either may be -1.

​
Outputs

​
Step by step

  1. Wire upstream video, Run upstream.
  2. Set width/height (e.g. 1280, 720).
  3. ▶ Run.
  4. If both dimensions are fixed and ratio differs from source, image stretches — use -1 on one axis to preserve aspect.
This page covers one node only. For end-to-end workflows, multi-stage pipelines, type conversion, and design rationale, see the ComfyTV user guides on GitHub:

​
Repository and workflows

​
FAQ

Q: Stretched picture?
A: Fixed width+height with wrong ratio stretches; try width=1280, height=-1. Q: vs Video Upscale?
A: Resize is geometric; Upscale (⏳ pending) is AI per-frame enlargement. Q: mesh2motion first?
A: → ComfyTV Video, then Resize.
  • Video Crop / Video Clip
  • Video Upscale (roadmap)