Prediction-Based Scale Adaptive Correlation Filter Tracker

Abstract

Although correlation filter-based trackers have demonstrated excellent performance for visual object tracking, there remain several challenges to be addressed. In this work, we propose a novel tracker based on the correlation filter framework. Traditional trackers face difficulty in accurately adapting to changes in the scale of the target when the target moves quickly. To address this, we suggest a scale adaptive scheme based on prediction scales. We also incorporate a speed-based adaptive model update method to further improve overall tracking performance. Experiments with samples from the OTB100 and KITTI datasets demonstrate that our method outperforms existing state-of-the-art tracking algorithms in fast motion scenes.

A Lightweight Object Detection Network for Real-Time Detection of Driver Handheld Call on Embedded Devices

“…Traffic congestion is a huge problem [ 1 ] facing the world in the process of rapid urbanization. In order to avoid traffic accidents, algorithms and embedded environment applications of drivers' dangerous behavior detection [ 2 ], such as handheld receiving and calling, have been studied extensively [ 3 – 7 ]. Most of the research on object detection networks aims at improving accuracy [ 8 ] but ignores the problems of model, calculation amount, and the number of parameters.…”