Beyond Correlation Filters: Learning Continuous Convolution Operators for Visual Tracking
This work addresses the problem of improving visual object tracking accuracy for applications like surveillance and robotics, though it is incremental as it builds upon the existing DCF framework.
The paper tackles the limitation of Discriminative Correlation Filters (DCF) to single-resolution feature maps in visual object tracking by introducing a novel formulation for training continuous convolution filters, resulting in superior performance with gains such as +5.1% in mean OP on OTB-2015 and a 20% relative reduction in failure rate on VOT2015.
Discriminative Correlation Filters (DCF) have demonstrated excellent performance for visual object tracking. The key to their success is the ability to efficiently exploit available negative data by including all shifted versions of a training sample. However, the underlying DCF formulation is restricted to single-resolution feature maps, significantly limiting its potential. In this paper, we go beyond the conventional DCF framework and introduce a novel formulation for training continuous convolution filters. We employ an implicit interpolation model to pose the learning problem in the continuous spatial domain. Our proposed formulation enables efficient integration of multi-resolution deep feature maps, leading to superior results on three object tracking benchmarks: OTB-2015 (+5.1% in mean OP), Temple-Color (+4.6% in mean OP), and VOT2015 (20% relative reduction in failure rate). Additionally, our approach is capable of sub-pixel localization, crucial for the task of accurate feature point tracking. We also demonstrate the effectiveness of our learning formulation in extensive feature point tracking experiments. Code and supplementary material are available at http://www.cvl.isy.liu.se/research/objrec/visualtracking/conttrack/index.html.