# DL Streamer Edge AI Extension

DL Streamer Edge AI Extension is a self-hosted ai app. Intel DL Streamer Edge AI Extension was a discontinued, archived project that provided Docker images extending the DL Streamer video analytics pipeline with OpenVINO-powered deep learning inference for Intel-based... It is open source and self-hosted, installable on Docker Hub.

Canonical page: https://ourlocalhost.com/app/6687/dl-streamer-edge-ai-extension

## Description

Intel DL Streamer Edge AI Extension was a discontinued, archived project that provided Docker images extending the DL Streamer video analytics pipeline with OpenVINO-powered deep learning inference for Intel-based workloads, but its maintenance has ended and it has known security escapes, making it unsuitable for new deployments.

## Platform availability

Available on 0 of 10 platforms.

## Health score

11/100 (Low), based on 3 of 4 factors.
- Maintained: 0 (GitHub repository is archived)
- Popular: 16 (5 GitHub stars)
- Easy to install: 25 (0 install platforms, install notes, Docker image)

## Links

- GitHub: https://github.com/dlstreamer/edge-ai-extension
- Docker Hub: https://hub.docker.com/r/intel/dlstreamer-edge-ai-extension

## Common questions

### Which inference accelerators are supported by the pre-built container?

The pre-built container supports CPU, Intel integrated GPU, Intel Neural Compute Stick 2, and HDDL-R inference accelerators.

### Which protocol APIs can clients use to send frames to the Edge AI Extension?

It exposes gRPC and HTTP protocol APIs. gRPC is intended for fast transfer and low latency, while HTTP returns inference results as JSON.

### What should I try if a DL Streamer installation on an Intel Edge AI Box (ESDQ) keeps failing?

If installation issues persist, use the new Edge software package on a fresh Ubuntu 22.04 system.

### How can I reduce first inference latency in Image Ingestor mode?

Use the REST API in Image Ingestor mode, because the troubleshooting docs state it has low first inference latency.

### What spatial analytics features are included?

It includes Object Line Crossing and Object Zone Count.

---
Source: Our Local Host (https://ourlocalhost.com/app/6687/dl-streamer-edge-ai-extension), the self-hosted and open-source app directory.
