/tflite_gles_app

several applications using TensorFlowLite for DNN inference and OpenGLES for visualization.

Primary LanguageCMIT LicenseMIT

gles_app

This repository contains several applications which invoke DNN inference with TensorFlow Lite and visualizes its result with OpenGLES.

applications

App name Descriptions
gl2detection img
Object Detection using MobileNet SSD.
gl2posenet img
Pose Estimation.
gl2segmentation img
Semantic image segmentation using Deeplab.

tested platforms

You can select the platform by editing Makefile.env.

  • Linux PC (XWindow/Wayland/DRM)
  • NVIDIA Jetson TX2 (XWindow/Wayland/EGLStream)
  • RaspberryPi (Dispmanx)

How to Build & Run

For Linux X11:

> make TARGET_ENV=x11
> cd gl2detection
> ./gl2detection

For Raspberry Pi (native build on RaaspberryPi)

> make TARGET_ENV=raspi
> cd gl2detection
> ./gl2detection