Web of Science: 13 citas, Scopus: 17 citas, Google Scholar: citas
Training a binary weight object detector by knowledge transfer for autonomous driving
Xu, Jiaolong (Centre de Visió per Computador (Bellaterra, Catalunya))
Nie, Yiming (Unmanned Systems Research Center (Beining, Xina))
Wang, Peng (Unmanned Systems Research Center (Beining, Xina))
López Peña, Antonio M. (Centre de Visió per Computador (Bellaterra, Catalunya))

Publicación: Institute of Electrical and Electronics Engineers (IEEE), cop. 2019
Descripción: 6 pàg.
Resumen: Autonomous driving has harsh requirements of small model size and energy efficiency, in order to enable the embedded system to achieve real-time on-board object detection. Recent deep convolutional neural network based object detectors have achieved state-of-the-art accuracy. However, such models are trained with numerous parameters and their high computational costs and large storage prohibit the deployment to memory and computation resource limited systems. Low-precision neural networks are popular techniques for reducing the computation requirements and memory footprint. Among them, binary weight neural networks (BWNs) are the extreme case which quantizes the float-point into just 1 bit. BWNs are difficult to train and suffer from accuracy deprecation due to the extreme low-bit representation. To address this problem, we propose a knowledge transfer (KT) method to aid the training of BWN using a full-precision teacher network. We built DarkNet- and MobileNet-based binary weight YOLO-v2 detectors and conduct experiments on KITTI benchmark for car, pedestrian and cyclist detection. The experimental results show that the proposed method maintains high detection accuracy while reducing the model size of DarkNet-YOLO from 257 MB to 8. 8 MB and MobileNet-YOLO from 193 MB to 7. 9 MB.
Ayudas: Agencia Estatal de Investigación TIN2017-88709-R
Nota: Altres ajuts: the authors want to acknowledge the Spanish DGT project SPIP2017-02237 and the Generalitat de Catalunya CERCA Program and its ACCIO agency
Derechos: Tots els drets reservats.
Lengua: Anglès
Documento: Capítol de llibre ; recerca ; Versió acceptada per publicar
Materia: SDG 7 - Affordable and Clean Energy ; Training ; Detectors ; Neural networks ; Quantization (signal) ; Knowledge transfer ; Task analysis ; Autonomous vehicles
Publicado en: 2019 International Conference on Robotics and Automation (ICRA), 2019, p. 2379-2384, ISBN 978-1-5386-6027-0

DOI: 10.1109/ICRA.2019.8793743


Postprint
7 p, 661.6 KB

El registro aparece en las colecciones:
Libros y colecciones > Capítulos de libros

 Registro creado el 2023-05-15, última modificación el 2023-06-02



   Favorit i Compartir