AI Inference Platform (AIP) consumes real-time media streams and executes AI-based worker-services (called processors) to enrich the media streams before propagating the media streams onwards to other consumers. This repository holds an open, protobuf and gRPC based API definition that processors implement to receive and reply with enrichments.
Compatible processors implement two gRPC services:
- configuration service: processors identify the kind and version of enrichment they perform
- processor service (v1 and v2): versioned processor implementations for receiving media frames and metadata and returning enrichments, such as AI inferences, georegistration, or computing tracks from prior inferences
Please contact us at aip-questions@palantir.com with any questions or feedback.
This repository is licensed under the Apache 2.0 License (https://www.apache.org/licenses/LICENSE-2.0).