ECDet: an efficient convolutional network for real-time object detection

Jie Wang; Quan Zhou; Dechun Cong; Xin Jin; Weihua Ou

doi:10.1117/12.2574632

12 October 2020 ECDet: an efficient convolutional network for real-time object detection

Jie Wang, Quan Zhou, Dechun Cong, Xin Jin, Weihua Ou

Proceedings Volume 11574, International Symposium on Artificial Intelligence and Robotics 2020; 1157402 (2020) https://doi.org/10.1117/12.2574632
Event: International Symposium on Artificial Intelligence and Robotics (ISAIR), 2020, Kitakyushu, Japan

Abstract

This paper introduces a lightweight convolutional neural network, called ECDet, for real-time accurate object detection. In contrast to recent advances of lightweight networks that prefer to use pointwise convolution for changing the number of feature map’s channel, ECDet makes an effort to design equal channel block for constructing the whole backbone network architecture. Meanwhile, we deploy depth-wise convolution to compress the feature pyramid network (FPN) detection head. The experiments show that ECDet only has 3.19 M model size and needs only 3.48B FLOPs with a 416×416 input image. Our method has a 5% improvement in accuracy compared to YOLO Nano, and it requires less computation. The comprehensive experiments demonstrate that our model achieves promising results in terms of available speed and accuracy trade-off on PASCAL VOC 2007 datasets.

Citation Download Citation

Jie Wang, Quan Zhou, Dechun Cong, Xin Jin, and Weihua Ou "ECDet: an efficient convolutional network for real-time object detection", Proc. SPIE 11574, International Symposium on Artificial Intelligence and Robotics 2020, 1157402 (12 October 2020); https://doi.org/10.1117/12.2574632

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available