> [!NOTE]
> The `main` branch is under active development and may be unstable. For production use, please pick a stable version from the [Releases](https://github.com/open-edge-platform/dlstreamer/releases) page.
<div align="center">
# Deep Learning Streamer (DL Streamer)
**Hardware-accelerated video analytics pipelines — CPU, GPU and NPU, from a single line of code to production-grade edge AI**
[](./LICENSE)
[](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/system_requirements.html)
[](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/system_requirements.html)
[](https://docs.openvino.ai)
[](https://gstreamer.freedesktop.org)
[](https://hub.docker.com/r/intel/dlstreamer)
[](https://github.com/open-edge-platform)
<img src="./hero.jpg" width="800" alt="DL Streamer sample outputs">
[Get Started](#quick-start---installation) • [Run Your Pipeline](#quick-start---run-your-pipeline) • [Samples](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/samples.html) • [Elements](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/elements/elements.html) • [Documentation](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/index.html) • [Contributing](./CONTRIBUTING.md)
</div>
---
## What is DL Streamer?
**DL Streamer** is an open-source media analytics framework built on [GStreamer](https://gstreamer.freedesktop.org). It lets you build video and audio intelligence pipelines — from a simple object detection command line to a multi-stream, multi-sensor production deployment, all with minimal code.
- Powered by **[OpenVINO™](https://docs.openvino.ai)** for optimized inference on Intel CPU, GPU, and NPU.
- Pipelines are described as **simple strings** (or Python/C++ code) and executed with full hardware acceleration.
- Ships with **30+ ready-to-run samples** covering detection, classification, tracking, VLMs, LiDAR and more.
- Part of the **[Intel Open Edge Platform](https://github.com/open-edge-platform)**.
---
## Why DL Streamer?
| Benefit | Details |
|---|---|
| **One-line pipelines** | Build a working detection pipeline in a single `gst-launch-1.0` command |
| **Hardware acceleration** | Targets CPU, GPU, and NPU on Intel platforms from a single codebase |
| **VLM & GenAI ready** | Run Vision-Language Models (MiniCPM-V, CLIP, Whisper) in a GStreamer pipeline |
| **GstAnalytics compliance** | Supports the GStreamer industry metadata standard for interoperability with other GStreamer-based tools |
| **Messaging integration** | Publish inference results directly to MQTT or Kafka with built-in elements — no extra code required |
| **Python-first extensibility** | Add custom logic as Python callbacks or full Python GStreamer elements — no C++ required |
| **Multi-stream, multi-sensor** | Mux/demux dozens of RTSP streams, LiDAR frames, and radar point clouds in one process |
| **Geti™, Ultralytics & HuggingFace support** | Deploy models from Geti™ Studio, Ultralytics, Hugging Face, or any ONNX/OpenVINO IR model directly |
<div align="center">
[](https://www.youtube.com/watch?v=1x7LTZhEadI&list=PLg-UKERBljNySiBeQaLtzF0yZGOmKrBRF)
</div>
---
## Quick Start - Installation
### Step 1 — Install GPU/NPU drivers (required for Docker and native install)
```bash
cd ~
wget https://raw.githubusercontent.com/open-edge-platform/dlstreamer/main/scripts/DLS_install_prerequisites.sh
chmod +x DLS_install_prerequisites.sh
./DLS_install_prerequisites.sh
```
> This script detects your Intel GPU/NPU, installs the correct drivers for Ubuntu 22.04 or 24.04, and adds your user to the required groups. Use `--reinstall-npu-driver=yes` to force-reinstall the NPU driver. Run `./DLS_install_prerequisites.sh --help` for all options.
### Step 2 — Install DL Streamer
**Option A — Docker (recommended, zero setup)**:
```bash
# Run once on the host to allow X11 forwarding from containers
xhost +local:docker
docker run -it --rm \
--device /dev/dri \
--group-add $(stat -c "%g" /dev/dri/render*) \
-e DISPLAY=$DISPLAY \
-e XDG_RUNTIME_DIR=/tmp \
-v /tmp/.X11-unix:/tmp/.X11-unix \
intel/dlstreamer:latest
```
> To use the NPU, also add `--device /dev/accel --group-add $(stat -c "%g" /dev/accel/accel*)` to the `docker run` command.
**Option B — Native install (Ubuntu 24.04)**:
```bash
sudo -E wget -O- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB | gpg --dearmor | sudo tee /usr/share/keyrings/intel-gpg-archive-keyring.gpg > /dev/null
sudo -E wget -O- https://apt.repos.intel.com/edgeai/dlstreamer/GPG-PUB-KEY-INTEL-DLS.gpg | sudo tee /usr/share/keyrings/dls-archive-keyring.gpg > /dev/null
echo "deb [signed-by=/usr/share/keyrings/dls-archive-keyring.gpg] https://apt.repos.intel.com/edgeai/dlstreamer/ubuntu24 ubuntu24 main" | sudo tee /etc/apt/sources.list.d/intel-dlstreamer.list
sudo bash -c 'echo "deb [signed-by=/usr/share/keyrings/intel-gpg-archive-keyring.gpg] https://apt.repos.intel.com/openvino ubuntu24 main" | sudo tee /etc/apt/sources.list.d/intel-openvino.list'
sudo apt update && sudo apt-get install -y intel-dlstreamer
```
Full installation guide: [Install Guide for Ubuntu](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/install/install_guide_ubuntu.html) | [Windows](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/install/install_guide_windows.html)
---
## Quick Start - Run Your Pipeline
**Step 1 — Set up models and environment** (inside the container or on a native install):
```bash
cd ~
curl -LsSf https://astral.sh/uv/install.sh | sh
uv venv .dls-venv && source .dls-venv/bin/activate
uv pip install -r /opt/intel/dlstreamer/scripts/download_models/requirements_download_ultralytics_models.txt
python3 /opt/intel/dlstreamer/scripts/download_models/download_ultralytics_models.py \
--model yolo11n.pt \
--outdir ~/models/yolo11n \
--int8
source /opt/intel/dlstreamer/scripts/setup_dls_env.sh
```
**Step 2 — Run the pipeline.** Change `device=GPU` to `device=CPU` or `device=NPU` — no other code changes needed.
```bash
gst-launch-1.0 \
urisourcebin buffer-size=4096 uri=https://videos.pexels.com/video-files/1192116/1192116-sd_640_360_30fps.mp4 ! \
decodebin3 ! \
gvadetect model=~/models/yolo11n/yolo11n.xml device=GPU ! \
queue ! \
gvawatermark ! \
gvafpscounter ! \
videoconvert ! autovideosink sync=false
```
Output to JSON (works everywhere, including headless Docker):
```bash
gst-launch-1.0 \
urisourcebin buffer-size=4096 uri=https://videos.pexels.com/video-files/1192116/1192116-sd_640_360_30fps.mp4 ! \
decodebin3 ! \
gvadetect model=~/models/yolo11n/yolo11n.xml device=GPU ! \
queue ! \
gvafpscounter ! \
gvametaconvert format=json ! \
gvametapublish file-format=json-lines file-path=output.json ! fakesink async=false
```
### Python API
Create a file `detect.py`:
```python
import gi
gi.require_version("Gst", "1.0")
from gi.repository import Gst
import os
Gst.init([])
video_url = "https://videos.pexels.com/video-files/1192116/1192116-sd_640_360_30fps.mp4"
model = os.path.expanduser("~/models/yolo11n/yolo11n.xml")
pipeline = Gst.parse_launch(f"""
urisourcebin buffer-size=4096 uri={video_url} !
decodebin3 !
gvadetect model={model} device=GPU !
queue !
gvafpscounter !
gvametaconvert format=json !
gvametapublish file-format=json-lines file-path=output_from_python.json ! fakesink async=false
""")
pipeline.set_state(Gst.State.PLAYING)
bus = pipeline.get_bus()
bus.timed_pop_filtered(Gst.CLOCK_TIME_NONE, Gst.MessageType.EOS | Gst.MessageType.ERROR)
pipeline.set_state(Gst.State.NULL)
```
Then run it and wait for results:
```bash
python3 detect.py
# Detection results are written to output_from_python.json as the pipeline processes frames
# Each line is a JSON object with detected objects, labels, and bounding boxes
cat output_from_python.json
```
> If any Python dependencies are missing, refer to the [Python dependencies install guide](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/dev_guide/advanced_install/advanced_install_guide_compilation.html#step-9-install-python-dependencies-optional).
---
## DL Streamer Elements
| Category | Key Elements |
|---|---|
| **Inference** | `gvadetect` · `gvaclassify` · `gvainference` · `gvagenai` · `gvaaudiotranscribe` |
| **Analytics** | `gvatrack` · `gvaanalytics` · `gvastreammux` / `gvastreamdemux` · `gvamotiondetect` |
| **Output** | `gvawatermark` · `gvametaconvert` · `gvametapublish` · `gvafpscounter` |
| **3D / Sensors** | `g3dlidarparse` · `g3dinference` · `g3dradarprocess` |
[Full elements reference →](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/elements/elements.html)
---
## Samples
30+ samples across Python, C++, and `gst-launch` command lines:
| Category | Samples |
|---|---|
| **Detection** | [YOLO detection](./samples/gstreamer/gst_launch/detection_with_yolo/README.md), [Face detection + classification](./samples/gstreamer/gst_launch/face_detection_and_classification/README.md), [Depth estimation](./samples/gstreamer/gst_launch/depth_estimation/README.md) |
| **Segmentation & Pose** | [Instance segmentation](./samples/gstreamer/gst_launch/instance_segmentation/README.md), [Human pose estimation](./samples/gstreamer/gst_launch/human_pose_estimation/README.md) |
| **Tracking** | [Vehicle & pedestrian tracking](./samples/gstreamer/gst_launch/vehicle_pedestrian_tracking/README.md), [Vehicle counter with tripwires](./samples/gstreamer/python/gvaanalytics_tripwire/README.md) |
| **VLM / GenAI** | [VLM video summarization](./samples/gstreamer/gst_launch/gvagenai/README.md), [VLM alerts](./samples/gstreamer/python/vlm_alerts/README.md), [VLM self-checkout](./samples/gstreamer/python/vlm_self_checkout/README.md) |
| **Multi-stream** | [Multi-camera deployment](./samples/gstreamer/gst_launch/multi_stream/README.md), [Stream mux/demux](./samples/gstreamer/gst_launch/stream_mux_and_demux/README.md) |
| **3D Sensors** | [LiDAR parsing](./samples/gstreamer/gst_launch/g3dlidarparse/README.md), [PointPillars 3D detection](./samples/gstreamer/gst_launch/g3dinference/README.md), [Radar processing](./samples/gstreamer/gst_launch/g3dradarprocess/README.md) |
| **Integration** | [ONVIF camera discovery](./samples/gstreamer/python/onvif_cameras_discovery/README.md), [Geti™ model deployment](./samples/gstreamer/gst_launch/geti_deployment/README.md), [Metadata to MQTT/Kafka](./samples/gstreamer/gst_launch/metapublish/README.md) |
| **Python extensibility** | [Custom Python GStreamer elements](./samples/gstreamer/gst_launch/python-elements/face_detection_and_classification/README.md), [Smart NVR with recording](./samples/gstreamer/python/smart_nvr/README.md)
[Browse all samples →](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/samples.html)
---
## Supported Platforms
| Hardware | CPU | GPU | NPU |
|---|:---:|:---:|:---:|
| Intel Core Ultra series 1–3 (Meteor / Lunar / Arrow / Panther Lake) | ✅ | ✅ | ✅ |
| Intel Arc discrete GPU (Alchemist, Battlemage) | — | ✅ | — |
| 11th–13th Gen Intel Core | ✅ | ✅ | — |
Operating systems: **Ubuntu 22.04 / 24.04**, **Windows 11**.
[Full system requirements →](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/system_requirements.html)
---
## Documentation
| Resource | Link |
|---|---|
| Overview | [Overview](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/index.html) |
| Elements Reference | [Elements Reference](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/elements/elements.html) |
| Supported Models | [Supported Models](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/supported_models.html) |
| Developer Guide | [Developer Guide](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/dev_guide/dev_guide_index.html) |
| API Reference | [API Reference](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/api_ref/api_reference.html) |
| Metadata Guide | [Metadata Guide](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dlstreamer/dev_guide/metadata.html) |
| Video Tutorials | [YouTube Playlist](https://www.youtube.com/playlist?list=PLg-UKERBljNySiBeQaLtzF0yZGOmKrBRF) |
---
## Contributing
We welcome contributions! Please read [CONTRIBUTING.md](./CONTRIBUTING.md) and follow the [Code Style Guide](./CODESTYLE.md).
For security issues, see [SECURITY.md](./SECURITY.md).
---
## License
DL Streamer is licensed under the [MIT License](./LICENSE).
---
<div align="center">
*Intel, the Intel logo, OpenVINO, OpenVINO logo, Intel Geti, Intel Core, Intel Arc, and Intel Iris are trademarks of Intel Corporation or its subsidiaries.*
*GStreamer is a trademark of the GStreamer project.*
</div>