SINet: A scale-insensitive convolutional neural network for fast vehicle detection

Publication Type

Journal Article

Publication Date

3-2019

Abstract

Vision-based vehicle detection approaches achieve incredible success in recent years with the development of deep convolutional neural network (CNN). However, existing CNN-based algorithms suffer from the problem that the convolutional features are scale-sensitive in object detection task but it is common that traffic images and videos contain vehicles with a large variance of scales. In this paper, we delve into the source of scale sensitivity, and reveal two key issues: 1) existing RoI pooling destroys the structure of small scale objects and 2) the large intra-class distance for a large variance of scales exceeds the representation capability of a single network. Based on these findings, we present a scale-insensitive convolutional neural network (SINet) for fast detecting vehicles with a large variance of scales. First, we present a context-aware RoI pooling to maintain the contextual information and original structure of small scale objects. Second, we present a multi-branch decision network to minimize the intra-class distance of features. These lightweight techniques bring zero extra time complexity but prominent detection accuracy improvement. The proposed techniques can be equipped with any deep network architectures and keep them trained end-to-end. Our SINet achieves state-of-the-art performance in terms of accuracy and speed (up to 37 FPS) on the KITTI benchmark and a new highway dataset, which contains a large variance of scales and extremely small objects.

Keywords

Fast object detection, Intelligent transportation system, Scale sensitivity, Vehicle detection

Discipline

Artificial Intelligence and Robotics

Research Areas

Software and Cyber-Physical Systems

Publication

IEEE Transactions on Intelligent Transportation Systems

Volume

20

Issue

3

First Page

1010

Last Page

1019

ISSN

1524-9050

Identifier

10.1109/TITS.2018.2838132

Publisher

Institute of Electrical and Electronics Engineers

Additional URL

https://doi.org/10.1109/TITS.2018.2838132

This document is currently not available here.

Share

COinS