IEEE Robotics & Automation Magazine - June 2020 - 157

13.3x

14

Speedup

12
10
8

5.8x

6
4
2
0

Figure 7. A typical output image of the DNN module with YOLO v3.

Edge
Detection

GPU

Embedded PC

3.51

2.63

0.23

46.6

17.2

2.14

running on the three types of computers used in the tests is
shown in Table 5.
In addition to absolute timings, it is illustrative to calculate
the speedup of the GPU with respect to the CPU: in other
words, how much faster than the CPU is the GPU for a given
application. The speedup results are shown in Figure 8, which
displays the values for each application (edge detection, ORB
features, optical flow, and DNNs) on each platform (desktop,
laptop, and embedded PC).
The overhead penalty can be noticed for the edge detection application on every platform. On the other hand, the
speedup is higher in the other applications, skyrocketing in
the last example (DNN for object recognition). This result is
not surprising, since CUDA is used intensively by the deeplearning community. But the benefits of using the GPU with
other OpenCV functions cannot be overlooked, obtaining
speedups of 580% and 290% for the computation of the optical flow in the desktop and laptop PCs, respectively.
OpenCV in ROS Projects
OpenCV is a widely used library in robotics projects; consequently, there is a precompiled module for the Robot Operating System (ROS) [14] (http://wiki.ros.org/opencv3); this,
unfortunately, does not include CUDA support. However,
GPU acceleration can still be used by replacing the standard
module with a CUDA-enabled version of the OpenCV
library. This can be done by installing the library and setting
the appropriate path in the file CMakeLists.txt of any
ROS module using OpenCV:
find_package(OpenCV REQUIRED
PATHS /usr/local
NO_DEFAULT_PATH
)

8
7
6
5
4
3
2
1
0

10
9
8
7
6
5
4
3
2
1
0

ORB
Optical Flow
Features
(a)

DNN

6.5x

2.9x
1.7x
0.6x
Edge
Detection

Speedup

CPU

Laptop PC

Speedup

Table 5. The frame rates (in hertz) for the YOLO
v3 network in the DNN module (higher, in bold,
is better).
Desktop PC

1.4x

0.8x

ORB
Optical Flow
Features
(b)

DNN

9.3x

1.5x

1.4x

0.2x
Edge
Detection

ORB
Optical Flow
Features
(c)
CPU

DNN

GPU

Figure 8. Performance comparisons of CPU versus GPU.
(a) Desktop PC. (b) Laptop PC. (c) Embedded PC.

In addition, all other ROS packages that are dependent upon
OpenCV (cv_bridge, image_pipeline, image_
transport, etc.) must be rebuilt; that is, their source code
must be downloaded into an ROS workspace and compiled
with catkin_make. A complete example of a simple subscriber is presented in the "ros" branch of the source code
repository of this article at https://github.com/RobInLabUJI/
opencv-cuda/tree/ros.
Converting an ROS topic image to a CUDA image is
straightforward; the topic message is converted to an
OpenCV image, and this image is uploaded to the GPU:
cv_ptr = cv_bridge::toCvCopy(msg, sensor_msgs::image_
encodings::BGR8);
gpuInImage.upload(cv_ptr->image);

JUNE 2020

*

IEEE ROBOTICS & AUTOMATION MAGAZINE

*

157
http://wiki.ros.org/opencv3 https://www.github.com/RobInLabUJI/opencv-cuda/tree/ros https://www.github.com/RobInLabUJI/opencv-cuda/tree/ros
IEEE Robotics & Automation Magazine - June 2020

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - June 2020
