The project associates an image, video, or screen recording with the markup, relationships, descriptions, and results of computer vision techniques. The saved data is used in Screph Code, supported external tools or Automation Runtime. Working with the code and executing the script are launched separately.
The mode determines the available object types and tools. General and GUI Automation support basic workflows; Industrial and UAV remain experimental.
The source can be a screen area, window, monitor, camera, network stream, image, video, or folder. Capture preserves the connection of the frame with the source and position on the time scale.
In Selector you create elements, features, areas of interest, annotations, and typed relationships. Layout changes can be undone and redone.
A single method, sequence, or graph of methods can create masks, outlines, detections, recognized text, metrics, and a log. Methods with models require a prepared environment and model files.
Preview, overlay, mask and metrics help you evaluate the result. In supported processes, a proposed change can be accepted, modified, rejected, or recalculated.
The project stores the format version, source images, markup, links, groups, descriptions and links to attached results. When loading, the structure and associated files are checked.
Screph Code and supported external tools receive selected task data. Automation Runtime uses the main project along with a separately prepared script.
Stores typed relationships between elements and features of the main markup.
A processing graph with typed inputs and outputs and an execution log. It transforms visual data.
Contains user actions and associated frames. Action CV saves suggested results next to the original entry.