Developer Guide

• Advanced Installation Guide

  • Ubuntu advanced installation - prerequisites

  • Advanced installation - compilation from source files

  • Ubuntu advanced installation - build Docker image

  • Ubuntu advanced uninstall guide

• Metadata

• Model Preparation

  • 1. Model file format used by OpenVINO™ Toolkit

  • 2. Model pre- and post-processing

  • 3. Specify model files in GStreamer elements

• OpenVINO Custom Operations

• Model Info Section

• Python Bindings

  • 1. GStreamer Python bindings

  • 2. Video-analytics specific Python bindings

  • 3. gvapython element

  • 4. Performance considerations

• Custom GStreamer Plugin Installation

  • 1. Install custom GStreamer plugin(s)

  • 2. Update GStreamer plugin(s) settings

• Custom Processing

  • 1. Consume tensor data and parse/convert it on application side

  • 2. Set C/Python callback in the middle of GStreamer pipeline

  • 3. Insert gvapython element and provide Python callback function

  • 4. Insert new GStreamer element implemented on C/C++

  • 5. Modify source code of post-processors for gvadetect/gvaclassify elements

  • 6. Create custom post-processing library

• Object Tracking

  • Object tracking types

  • Additional configuration

  • Sample

  • How to read object unique id

  • Performance considerations

• GPU device selection

  • 1. Media (VAAPI based) elements

  • 2. Inference (OpenVINO™ based) elements

  • 3. Media and Inference elements for GStreamer 1.24.0 and later versions

• Performance Guide

  • 1. Media and AI processing (single stream)

  • 2. Multi-stage pipeline with gvadetect and gvaclassify

  • 3. Multi-stream pipelines with single AI stage

  • 4. Multi-stream pipelines with multiple AI stages

  • 5. Multi-stream pipelines with meta-aggregation element

  • 6. The Intel® DL Streamer Pipeline Framework performance benchmark results

• Profiling with Intel VTune™

  • 1. Install VTune™

  • 2. Configure VTune™ host platform, Windows-to-Linux remote profiling method

  • 3. Results Analysis

• Converting NVIDIA DeepStream Pipelines to Intel® Deep Learning Streamer (Intel® DL Streamer) Pipeline Framework

  • Contents

  • Preparing Your Model

  • Configuring Model for Intel® DL Streamer

  • GStreamer Pipeline Adjustments

  • Multiple Input Streams

  • DeepStream to DLStreamer Elements Mapping Cheetsheet

• How to Contribute

  • Coding Style

• Latency Tracer

  • Pipeline

  • Pipeline latency flag

  • Element latency flag

  • Setting interval example

• Model-proc File (legacy)

  • Contents

  • Overview

  • Pre-processing description (input_preproc)

  • Post-processing description (output_postproc)