Stereolabs ZED - YOLO 3D

This package lets you use YOLO (v2, v3 or v4), the deep learning object detector using the ZED stereo camera in Python 3 or C++.

Update : the ZED is now natively supported in YOLO !

1. Setup

Prerequisites

Windows 7 10, Ubuntu LTS, L4T
ZED SDK and its dependencies (CUDA)
OpenCV built with CUDA and cuDNN

Preparing OpenCV installation

cuDNN

In order to get the best performance, cuDNN should be install before building OpenCV. Heads over to this TensorFlow documentation article which explains how to setup both CUDA and cuDNN on Ubuntu and Windows.

OpenCV

OpenCV binaries can be downloaded and install from opencv.org.

Alternatively, on Ubuntu :

sudo apt install pkg-config libopencv-dev

CMake

On Windows, download and install CMAKE using the binary available here.

On Ubuntu, cmake can be installed using the package manager, i.e : sudo apt install cmake

However the default version of cmake might be too old, it can easily be updated using the script (located in this repository):

sudo bash cmake_apt_update.sh

ZED Support Using CMake (recommended)

If the ZED SDK is installed, CMake will automatically detect it and compile with the ZED support. During the CMake configuration, a message will confirm that the ZED SDK was found.

...
-- A library with BLAS API found.
-- ZED SDK enabled
-- Found OpenMP_C: -fopenmp (found version "4.5")
...

2. Launching the YOLO 3D in C++

Download the yolo weights, yolov4 for instance, and put them in the local folder.

cd zed_cpp_sample/

mkdir build
cd build
cmake ..
make

./build/yolo_zed

3. Launching the YOLO 3D in Python