FuCoLoT -- A Fully-Correlational Long-Term Tracker

TL;DR

FuCoLoT is a novel long-term tracking method that uses multiple correlation filters and a new detection mechanism to achieve state-of-the-art results with low memory and real-time speed.

Contribution

It introduces a fully correlational long-term tracker with a novel filter learning method and a new failure detection mechanism, outperforming existing methods on standard benchmarks.

Findings

Achieves state-of-the-art results on short-term benchmarks.

Outperforms current best on UAV20L long-term benchmark by over 19%.

Operates at 15fps with a small memory footprint.

Abstract

We propose FuCoLoT -- a Fully Correlational Long-term Tracker. It exploits the novel DCF constrained filter learning method to design a detector that is able to re-detect the target in the whole image efficiently. FuCoLoT maintains several correlation filters trained on different time scales that act as the detector components. A novel mechanism based on the correlation response is used for tracking failure estimation. FuCoLoT achieves state-of-the-art results on standard short-term benchmarks and it outperforms the current best-performing tracker on the long-term UAV20L benchmark by over 19%. It has an order of magnitude smaller memory footprint than its best-performing competitors and runs at 15fps in a single CPU thread.