IEEE Robotics & Automation Magazine - June 2020 - 154

Table 2. The computation times (in milliseconds)
for the edge detection algorithm (lower, in bold,
is better).
Desktop PC

Figure 4. The output image of the edge detection example.

Table 1. The technical specifications of the
systems used in the experiments.
GPU Features

Desktop PC

Laptop PC

Embedded
PC

CUDA cores

2,560

640

128

Memory

8 GB

4 GB

Memory interface GDDR5

GDDR5

LPDDR4

Memory interface 256 b
width

128 b

64 b

Memory bandwidth

320 GB/s

112 GB/s

25.6 GB/s

Power consumption

180 W

40-50 W

10 W

CPU memory. Figure 4 depicts the output for a frame of a
video recorded during a car navigation task.
To measure the average computing times of the algorithm,
we processed the frames of a benchmarking video on three
different hardware configurations of a CPU and GPU:
● Desktop PC, with a CPU Intel Core i7-6700 at 3.4 GHz and
a GPU GeForce GTX 1080
● Laptop PC, with a CPU Intel Core i7-8550U at 3.3 GHz
and a GPU GeForce GTX 1050
● Embedded PC, an NVIDIA Jetson Nano with an ARM-A57
processor and an integrated GPU.
The video consists of
50-s footage from a car
in a highway available at
It is challenging, yet
Udacity's Advanced Line
Finding Project (https://
potentially very rewarding,
g it hub.com/udacit y/
C a rN D -Adv a nc e dto accelerate OpenCV on
Lane-Lines), recorded at
25 Hz with a resolution
graphics processors.
of 1,280 x 720 24-b red,
green, and blue. The
main specifications of
the GPUs for the three systems are presented in Table 1.
The desktop PC features the most powerful CPU both in
terms of processing cores and transfer speed, but it also
154

IEEE ROBOTICS & AUTOMATION MAGAZINE

JUNE 2020

Laptop PC

Embedded PC

CPU

3.514 ± 0.272 4.562 ± 0.331

12.985 ± 1.197

GPU

4.422 ± 0.150

54.064 ± 1.993

7.469 ± 0.106

requires more energy compared to the laptop and embedded PCs, which are adequate for mounting on a small
robotic platform.
The source code with instructions for compilation and
execution is publicly available (https://github.com/
RobInLabUJI/opencv-cuda). For the sake of reproducibility,
we use docker (https://www.docker.com), a Linux container technology that offers some advantages for an easy replication of code: encapsulation, isolation, portability, and
control. In addition, containers have less overhead than virtual machines, and they can access the GPU transparently
(usually the impact is on the order of less than 1% and
hardly noticeable). As a downside, the GPU-enabled version of docker (Nvidia-docker) does not yet support Windows or macOS.
The code can also be compiled and executed natively in a
Linux computer (as long as all the requirements are previously installed-basically, OpenCV and CUDA) with the typical
building commands:
mkdir -p build
cd build
cmake . .
make

The results are shown in Table 2. They measure the
mean and standard deviation of the execution time for
1,200 frames in the video (the initial 60 frames are
skipped to avoid initialization delays). The execution
time is measured starting from the first call to processing functions and continues until the final result is
returned; this result is averaged with a moving window
of 30 frames. For the GPU cases, the measured time
includes the uploading of the initial image to GPU
memory and the downloading of the result image back
to CPU memory. Visual information [edges, object
request broker (ORB) keypoints, and optical flow] is
included in the measured code for clarity and debugging
purposes, although in a real setup it could be removed to
increase the throughput.
It is worth noting that, for this application, the CPUs are
faster than the GPUs in all three systems; edge detection is a
relatively simple computation, and the execution time is small
compared with the overhead of transferring the images into
the GPU memory.

https://www.github.com/robInLabUJI/opencv-cuda https://www.github.com/robInLabUJI/opencv-cuda https://www.docker.com http://www.github.com/udacity/Ca rND-Advanced-Lane-Lines http://www.github.com/udacity/Ca rND-Advanced-Lane-Lines http://www.github.com/udacity/Ca rND-Advanced-Lane-Lines http://www.github.com/udacity/Ca rND-Advanced-Lane-Lines

IEEE Robotics & Automation Magazine - June 2020

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