Learning Mobile CNN Feature Extraction Toward Fast Computation of Visual Object Tracking

TL;DR

This paper introduces a lightweight, high-speed CNN-based object tracking method optimized for low-resource environments, utilizing MobileNetV3 and feature selection to achieve high accuracy without online learning.

Contribution

It proposes a novel architecture using MobileNetV3 and feature map selection for efficient, high-precision object tracking suitable for low-resource devices.

Findings

Achieves high tracking accuracy on Visual Tracker Benchmark

Operates efficiently with no online learning required

Suitable for low computational resource environments

Abstract

In this paper, we construct a lightweight, high-precision and high-speed object tracking using a trained CNN. Conventional methods with trained CNNs use VGG16 network which requires powerful computational resources. Therefore, there is a problem that it is difficult to apply in low computation resources environments. To solve this problem, we use MobileNetV3, which is a CNN for mobile terminals.Based on Feature Map Selection Tracking, we propose a new architecture that extracts effective features of MobileNet for object tracking. The architecture requires no online learning but only offline learning. In addition, by using features of objects other than tracking target, the features of tracking target are extracted more efficiently. We measure the tracking accuracy with Visual Tracker Benchmark and confirm that the proposed method can perform high-precision and high-speed calculation even in low computation resource environments.