Use GPU ORB Extractor with OpenCV-free Library#

This tutorial demonstrates how to use the GPU orb-extractor feature OpenCV-free library. The GPU orb-extractor feature OpenCV-free library provides similar features, except input and output structures are defined within this library.

Tutorial#

Prepare the environment:
```
cd /opt/intel/orb_lze/samples/
```
main.cpp should be in the directory. View it on GitHub to read the comments for the code.

Note: Refer to the explanations in the Use the GPU ORB Extractor tutorial for details on how to use the orb-extractor feature library API.

Build the code:

cp -r /opt/intel/orb_lze/samples/ ~/orb_lze_samples
cd ~/orb_lze_samples/
mkdir build
cd build
cmake -DBUILD_OPENCV_FREE=ON ../
make -j$(nproc)

Run the binary:

./feature_extract -h

Available command line arguments:

Usage: ./feature_extract --images=<> --image_path=<> --threads=<>

  --images <integer>     : Number of images or number of cameras. Default value: 1
  --image_path <string>  : Path to input image files. Default value: image.jpg
  --threads <integer>    : Number of threads to run. Default value: 1
  --iterations <integer> : Number of iterations to run. Default value: 10

The following command runs four threads, each thread taking two camera image inputs:

./feature_extract --images=2 --threads=4

Expected results example:
```
./feature_extract --images=2 --threads=4
 iteration 10/10
 Thread:2: gpu host time=21.4233
 iteration 10/10
 Thread:1: gpu host time=21.133
 iteration 10/10
 Thread:4: gpu host time=20.9086
 iteration 10/10
 Thread:3: gpu host time=20.6155
```
After execution, the input image displays keypoints as blue dots.

Note: You can specify the number of images per thread and the number of threads to execute. You can process multiple image inputs within a single thread of the extract API, or process one or more image inputs using multiple threads with extract API calls.

Use GPU ORB Extractor with OpenCV-free Library#

Tutorial#

This Page