Installing GstInference

🚧 Documentation under development

The Getting started with AI on NXP i.MX8M Plus guide is currently under active development. Some sections may be incomplete or change without notice.

Questions? Contact RidgeRun or email to support@ridgerun.com.

Getting started with AI on NXP i.MX8M Plus
Introduction
Overview i.MX8M Plus hardware specifications Development tools Development kit DART-MX8M-PLUS Development kit Quick getting started Yocto minimal image Flashing SD card First steps on the board
Development
Gatesgarth and Hardknott Yocto versions Downloading Yocto images for NXP based BSP Building an imx-core-image Flashing SD card Integrating multimedia software stack Installing GStreamer Installing OpenCV Integrating Artificial Intelligence software stack Installing TensorFlow Lite for NXP based images Installing R2Inference Installing GstInference Developing software for the board Building the SDK Crosscompiling apps for GStreamer, TensorFlow Lite, and OpenCV Exploring TensorFlow Lite delegates for prototyping
Neural Processing Unit
Introduction Profiling the NPU Installing gputop Setting the board on debug mode Use Case experiments: Smart Parking Introduction to the use case License Plate Detection TinyYOLO version 3 Training a model compatible with GstInference License Plate Recognition Rosetta Getting the model prototyping Bash scripts for CPU usage and time estimation Serial experiments Parallel Experiments Results
GStreamer
Exploring Rosetta element for character recognition Example pipelines Smart Parking Video inferences Image inferences Video streaming
Contact Us

Configuring GstInference

Since GstInference requires R2Inference for a proper build, then you should follow the Installing R2Inference before going through this wiki page.

Please find the local.conf file at:

cd $HOME/<your image folder>/<your build directory>/conf/

Finally, add the following line at this local.conf file:

IMAGE_INSTALL_append += "gst-inference"

After all please ensure that the recipe is like the following:

SUMMARY = "GstInference"
DESCRIPTION = "A GStreamer deep learning inference framework"
HOMEPAGE = "https://developer.ridgerun.com/wiki/index.php?title=GstInference"
SECTION = "multimedia"
LICENSE = "LGPL2.1"

LIC_FILES_CHKSUM = "file://COPYING;md5=b5ec61ada91a1aad4812666edfd8c57e"

DEPENDS = "gstreamer1.0 gstreamer1.0-plugins-base r2inference opencv"

SRCBRANCH ?= "master"
SRCREV = "55cad49398398f1fcc424f76244e65585c7cd8f4"
SRC_URI = "git://github.com/RidgeRun/gst-inference;protocol=https;branch=${SRCBRANCH}"

EXTRA_OEMESON += " -Denable-gtk-doc=false"

S = "${WORKDIR}/git"

FILES_${PN} += "${libdir}/gstreamer-1.0/libgstinferenceoverlay.so "
FILES_${PN} += "${libdir}/gstreamer-1.0/libgstinference.so "
FILES_${PN} += "${libdir}/gstreamer-1.0/libgstinferenceutils.so "

inherit meson pkgconfig gettext

Building the image with the changes in local.conf and bblayers.conf

cd $HOME/<your image folder>

source setup-environment <your build directory>

bitbake imx-image-core

Testing the GstInference installation

Ensure that GstInference is integrated in GStreamer

After flashing the image into the SD with the changes described here, then we can test it in the i.MX8M Plus.

In the board, please execute the gst-inspect command, as follows:

gst-inspect-1.0 inference

This command will show something like the following output:

Plugin Details:
  Name                     inference
  Description              Infer pre-trained model on incomming image frames on a variety of architectures and different backends
  Filename                 /usr/lib/gstreamer-1.0/libgstinference.so
  Version                  0.11.0.1
  License                  LGPL
  Source module            gst-inference
  Binary package           GstInference
  Origin URL               Unknown package origin

  resnet50v1: resnet50v1
  inceptionv1: inceptionv1
  inceptionv2: inceptionv2
  mobilenetv2: mobilenetv2
  inceptionv3: inceptionv3
  inceptionv4: inceptionv4
  tinyyolov2: tinyyolov2
  tinyyolov3: tinyyolov3
  mobilenetv2ssd: mobilenetv2ssd
  Rosetta: rosetta

  10 features:
  +-- 10 elements

Executing some inferences

For this section, please refer to the Video inferences section.