Model-free Tracking with Deep Appearance and Motion Features Integration

TL;DR

This paper introduces AMNet, a real-time, model-free object tracking framework that integrates appearance and motion features using deep CNNs, achieving high accuracy without prior object information.

Contribution

It presents a novel end-to-end two-stream CNN architecture that combines appearance and motion cues for anonymous object tracking, advancing model-free tracking methods.

Findings

Achieves state-of-the-art performance on OTB and VOT benchmarks.

Operates in real-time with favorable speed.

Effectively tracks anonymous objects without prior object models.

Abstract

Being able to track an anonymous object, a model-free tracker is comprehensively applicable regardless of the target type. However, designing such a generalized framework is challenged by the lack of object-oriented prior information. As one solution, a real-time model-free object tracking approach is designed in this work relying on Convolutional Neural Networks (CNNs). To overcome the object-centric information scarcity, both appearance and motion features are deeply integrated by the proposed AMNet, which is an end-to-end offline trained two-stream network. Between the two parallel streams, the ANet investigates appearance features with a multi-scale Siamese atrous CNN, enabling the tracking-by-matching strategy. The MNet achieves deep motion detection to localize anonymous moving objects by processing generic motion features. The final tracking result at each frame is generated by fusing the output response maps from both sub-networks. The proposed AMNet reports leading performance on both OTB and VOT benchmark datasets with favorable real-time processing speed.