Real-Time Siamese Multiple Object Tracker with Enhanced Proposals

TL;DR

SiamMOTION is a real-time multiple object tracker that uses a novel proposal engine with attention mechanisms and a Siamese comparison head, achieving state-of-the-art performance on multiple benchmarks.

Contribution

It introduces a new proposal engine and comparison architecture that enable real-time tracking of many objects with high accuracy.

Findings

Achieves leading performance on five public benchmarks.

Operates in real-time with dozens of objects.

Outperforms existing methods in accuracy and efficiency.

Abstract

Maintaining the identity of multiple objects in real-time video is a challenging task, as it is not always feasible to run a detector on every frame. Thus, motion estimation systems are often employed, which either do not scale well with the number of targets or produce features with limited semantic information. To solve the aforementioned problems and allow the tracking of dozens of arbitrary objects in real-time, we propose SiamMOTION. SiamMOTION includes a novel proposal engine that produces quality features through an attention mechanism and a region-of-interest extractor fed by an inertia module and powered by a feature pyramid network. Finally, the extracted tensors enter a comparison head that efficiently matches pairs of exemplars and search areas, generating quality predictions via a pairwise depthwise region proposal network and a multi-object penalization module. SiamMOTION has been validated on five public benchmarks, achieving leading performance against current state-of-the-art trackers. Code available at: https://github.com/lorenzovaquero/SiamMOTION