Accelerating real-time embedded scene labeling with convolutional networks

Today there is a clear trend towards deploying advanced computer vision (CV) systems in a growing number of application scenarios with strong real-time and power constraints. Brain-inspired algorithms capable of achieving record-breaking results combined with embedded vision systems are the best can...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings - ACM IEEE Design Automation Conference S. 1 - 6
Hauptverfasser: Cavigelli, Lukas, Magno, Michele, Benini, Luca
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.06.2015
Schlagworte:
Accelerator
Convolution
Convolutional Networks
Feature extraction
Graphics processing units
Labeling
Performance evaluation
Real-time systems
Scene Labeling
Throughput
ISSN:0738-100X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Today there is a clear trend towards deploying advanced computer vision (CV) systems in a growing number of application scenarios with strong real-time and power constraints. Brain-inspired algorithms capable of achieving record-breaking results combined with embedded vision systems are the best candidate for the future of CV and video systems due to their flexibility and high accuracy in the area of image understanding. In this paper, we present an optimized convolutional network implementation suitable for real-time scene labeling on embedded platforms. We show that our algorithm can achieve up to 96GOp/s, running on the Nvidia Tegra K1 embedded SoC. We present experimental results, compare them to the state-of-the-art, and demonstrate that for scene labeling our approach achieves a 1.5x improvement in throughput when compared to a modern desktop CPU at a power budget of only 11 W.
ISSN:0738-100X
DOI:10.1145/2744769.2744788