pipeline
Pipeline-level metadata. All fields are optional.
streams
Define one or more input sources. The stream type is auto-detected from the URI.
models
The models list defines the AI inference pipeline. It can contain detectors, pose estimators, segmentation models, classifiers, Triton models, and an optional tracker. Entries are processed in the order they appear, which controls the inference chain.
sinks
Configure output destinations. You can define multiple outputs simultaneously (e.g., display + RTSP + Kafka).