Implementation and optimization of human tracking system using ARM embedded platform

Teoh, Shen Khang; Yap, Vooi Voon; Soh, Chit Siang; Sebastian, Patrick

doi:10.1109/icias.2012.6306217

Cited by 6 publications

(1 citation statement)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditionally, in the embedded world parallelism has been exploited by means of special-purpose DSPs relying on SIMD or VLIW. Two examples are the Qualcomm Hexagon DSP [39] that accelerates a Snapdragon 800 with VLIW DSPs and is effective for vision and context inference tasks [40], as well as the Neon SIMD extensions that are integrated in many ARM cores [41]. All these platforms, however, are not meant to couple with a low power microcontroller, as they are designed for high end embedded architectures with DRAM, memory management and complex operating systems with power budget in the hundreds of milliwatts at chip level, up to a few watts at system level.…”

Section: Related Workmentioning

confidence: 99%

Accelerated Visual Context Classification on a Low-Power Smartwatch

Conti

Palossi

Andri

et al. 2016

IEEE Trans. Human-Mach. Syst.

View full text Add to dashboard Cite

Data produced by wearable sensors is key in contexts such as performance enhancement and training help for sports and fitness, continuous monitoring for aging people and for chronic disease management, and in gaming and entertainment. Unfortunately, wearable devices currently on the market are either incapable of complex functionality or severely impaired by short battery lifetime. In this work, we present a smartwatch platform based on an ultra-low power (ULP) heterogeneous system composed by a TI MSP430 microcontroller, the PULP programmable parallel accelerator and a set of ULP sensors, including a camera. The embedded PULP accelerator enables state-of-the-art context classification based on Convolutional Neural Networks (CNNs) to be applied within a sub-10mW system power envelope. Our methodology enables to reach high accuracy in context classification over 5 classes (up to 84%, with 3 classes over 5 reaching more than 90% accuracy), while consuming 2.2mJ per classification, or an ultra-low energy consumption of less than 91uJ per classification with an accuracy of 64%-3.2× better than chance. Our results suggest that the proposed heterogeneous platform can provide up to 500× speedup with respect to the MSP430 within a similar power envelope, which would enable complex computer vision algorithms to be executed in highly power-constrained scenarios.

show abstract

Section: Related Workmentioning

confidence: 99%

Accelerated Visual Context Classification on a Low-Power Smartwatch

Conti

Palossi

Andri

et al. 2016

IEEE Trans. Human-Mach. Syst.

View full text Add to dashboard Cite

show abstract

Enhanced TLD-based video object-tracking implementation tested on embedded platforms

Otoom

Al-Louzi

2021

J Real-Time Image Proc

View full text Add to dashboard Cite

Real time industrial application of single board computer based color detection system

Solak

Bolat

2013

2013 8th International Conference on Electrical and Electronics Engineering (ELECO)

View full text Add to dashboard Cite

In this study, real time industrial application of single board computer based color detection system is realized. In this system, BeagleBoard-xM as a single board computer, a USB camera, a conveyor belt and an LCD7 touch screen are used. OpenCV is used as an image processing library in this color detection system. The main goal of this study is to define the number of different colored packages passing on the conveyor belt according to their color. Then, real time results of the number of the packages and the total package number are displayed on the LCD7 touch screen. At the same time, the USB camera image of the related package on the conveyor belt is monitorized on the same touch screen. If no image of any packages is taken by the USB camera during 60 seconds, the system is turned off.

show abstract

Implementation and optimization of human tracking system using ARM embedded platform

Cited by 6 publications

References 1 publication

Accelerated Visual Context Classification on a Low-Power Smartwatch

Accelerated Visual Context Classification on a Low-Power Smartwatch

Enhanced TLD-based video object-tracking implementation tested on embedded platforms

Real time industrial application of single board computer based color detection system

Contact Info

Product

Resources

About