ECO: Efficient Convolution Operators for Tracking
This work addresses the trade-off between speed and performance in visual object tracking, offering significant improvements for real-time applications, though it is incremental as it builds on existing DCF methods.
The paper tackles the computational complexity and over-fitting issues in Discriminative Correlation Filter (DCF) based tracking methods by introducing a factorized convolution operator, a compact generative model, and a conservative update strategy, resulting in a 20-fold speedup with deep features and a 13.0% relative gain in Expected Average Overlap on VOT2016, and a fast variant operating at 60 Hz with 65.0% AUC on OTB-2015.
In recent years, Discriminative Correlation Filter (DCF) based methods have significantly advanced the state-of-the-art in tracking. However, in the pursuit of ever increasing tracking performance, their characteristic speed and real-time capability have gradually faded. Further, the increasingly complex models, with massive number of trainable parameters, have introduced the risk of severe over-fitting. In this work, we tackle the key causes behind the problems of computational complexity and over-fitting, with the aim of simultaneously improving both speed and performance. We revisit the core DCF formulation and introduce: (i) a factorized convolution operator, which drastically reduces the number of parameters in the model; (ii) a compact generative model of the training sample distribution, that significantly reduces memory and time complexity, while providing better diversity of samples; (iii) a conservative model update strategy with improved robustness and reduced complexity. We perform comprehensive experiments on four benchmarks: VOT2016, UAV123, OTB-2015, and TempleColor. When using expensive deep features, our tracker provides a 20-fold speedup and achieves a 13.0% relative gain in Expected Average Overlap compared to the top ranked method in the VOT2016 challenge. Moreover, our fast variant, using hand-crafted features, operates at 60 Hz on a single CPU, while obtaining 65.0% AUC on OTB-2015.