Local Edge AI. Open Integration. Built for Real-World Deployment.
DSDevices Edge AI Service is a forthcoming open-source platform designed to make computer vision simple to deploy on DSDevices hardware.
Using a standard USB camera, supported DSDevices players can process visual data locally and make detection results available through a simple REST API for use by CMS platforms, integrators and third-party applications.
- Local computer-vision processing
- Open REST API
- Open-source GitHub release
- Support for DSDA1 and DSDX5
- CPU-based inference on current hardware
- Future NPU-accelerated support
- Custom AI development and integration services
Simple Local Architecture
The first release is being developed for the DSDA1 and DSDX5, using CPU-based inference on the current generation of devices.
USB Camera → DSDA1 / DSDX5 → DSDevices Edge AI Service → REST API → CMS / Third-Party App
Processing takes place locally on the device, reducing reliance on cloud-based AI services and giving developers a straightforward way to integrate computer-vision data into their own applications.
Designed for Easy Integration
DSDevices Edge AI Service will expose a simple REST API designed for use by CMS providers, integrators and developers.
Initial endpoints will include:
/api/status– service and processing status/api/detections– structured object-detection results/api/snapshot– current camera snapshot/api/csv– detection data in CSV format
Third-party applications can consume this output without needing to manage the underlying AI inference engine.
Open Source on GitHub
The project is planned for release on GitHub under the AGPL-3.0 licence, giving developers access to the source code and a practical reference implementation for building their own integrations.
The initial release will use YOLO-based object detection, while the DSDevices REST API will provide the integration layer.
This approach is intended to allow future AI models and processing methods to be introduced without changing how third-party applications connect.
Built for CMS Providers, Integrators and Developers
DSDevices Edge AI Service is being developed to support a broad range of computer-vision applications.
Potential use cases include:
- Object and people detection
- Occupancy and footfall monitoring
- Retail and display monitoring
- Event-based content triggering
- Operational monitoring
- Custom computer-vision workflows
CMS providers can integrate detection data directly into their platforms, while integrators and developers can build their own applications around the same open API.
Custom AI Development Services
Alongside the open-source platform, DSDevices will offer commercial development and integration services for customers requiring tailored solutions.
These services may include:
- Custom model development and training
- Dataset preparation and annotation
- Model optimisation
- CPU and NPU deployment optimisation
- API and application integration
- Deployment engineering
- Ongoing model and technical support
Customers will be able to start with the open platform and move to a customised AI solution where required.
Existing Hardware. New Possibilities.
The first release of DSDevices Edge AI Service is being developed for the existing DSDA1 and DSDX5 platforms using CPU-based processing.
Future DSDevices hardware will extend this capability with NPU acceleration, supporting higher-performance edge AI workloads and more demanding applications.
This allows the Edge AI Service to develop alongside the DSDevices hardware platform while maintaining a consistent integration layer for third-party applications.
Coming Soon
DSDevices Edge AI Service is currently in development.
Further information, developer documentation and the GitHub repository will be published as the project moves towards release.
Interested in integrating Edge AI with your CMS or application?
