Video
Updated: 27 Jan 2026
Video
Introduction #
Video nodes are nodes that are used to input, process and combine a variety of video sources. The Video Loader is used to load in video and image files, Video Capture is used to bring in live video capture feeds such as capture cards and webcams, and nodes such the Face Tracker and NVIDIA AR Body Tracker allow you to extract tracking data from a video source.
Pages #
Depth Processing
Depth Processing nodes generate and change depth data using various techniques.
| Depth Camera Colour Key | TBC |
| Kinect Surface Mask | Creates a image representing surface of the kinect data. |
| Luminance From Depth | Generates a greyscale luminance image from a depth source, usually a Depth Camera / Kinect Source |
| Stereo Image Rectification | Modify a stereo image to fix lineup. |
Feature Tracking
Feature tracking nodes detect and track features in an image.
| Blob Tracker | This node generate blobs from an input image, and outputs transformation information for each of those blobs. |
| Face Tracker | Searches the supplied image or video for faces. |
| NVIDIA AR Body Tracker | Utilise the NVIDIA Broadcast SDK on RTX GPUs to generate rigged skeleton poses based on 2D video. |
Image Processing
These nodes analyse and modify the image with different effects.
| Composite Sources | Composite two image sources to one image. |
| Downsample | Downsample the resolution of the input image. |
| Frame Buffer | Buffers a frame for a period of time. |
| Frame Delay Frame | Delays the image played by a specified frame amount. |
| Generate Normal Map | Generates a normal map based off the luminance values from an image. |
| Resize | Resizes an image to specified size. |
| Upsample | Upscale the resolution of the image. |
Input Output
These nodes control where video is imported from, and where video is output to.
| Colour And Depth Source | Combines two input images and generates a depth camera source from them. |
| Depth Camera Source | Read colour and depth data from a Kinect camera. |
| Dynamic Image Loader | Loads an image from disk based on a specified filename. |
| GBuffer To Image | Allows GBuffer channels from a rendered scene to be used as video nodes for compositing or export purposes. |
| Image Tile File Loader | Loads a series of images from disk and composes them as tiles inside a single large texture. |
| Multiplex Sources | Takes multiple video inputs and outputs a single video based on the Source Index attribute. |
| Rendered Video Source | Output the rendered image from a Render Layer node for other nodes to use. |
| Video NDI Source | Use input from an NDI IP video stream. |
Motion Processing
TODO Section Description
| Optical Flow | Analyses an image and visualise the movement of objects in the scene. |
| Temporal Filter | Combines multiple frames together and filters out overlapping values. |
Segmentation
TODO Section Description
| Background Subtract | TBC |
Nodes
| Video Capture | Use input from a video capture device (DirectShow / Media Foundation). |
| Video Clips | Transitions between video clips in the timeline. |
| Video Loader | Source for a video or image from a file or from a media server (when running as a block). |
| Video Null | Creates a copy of the upstream video node. |