DCCO: Towards Deformable Continuous Convolution Operators

TL;DR

This paper introduces a deformable convolution filter for object tracking that adapts to non-rigid target transformations, improving performance over traditional rigid models and achieving results comparable to state-of-the-art methods.

Contribution

It presents a unified deformable convolution filter formulation that jointly infers sub-filter coefficients and locations, addressing non-rigid transformations in tracking.

Findings

Improves baseline tracking performance.

Achieves results comparable to state-of-the-art.

Validated on multiple challenging benchmarks.

Abstract

Discriminative Correlation Filter (DCF) based methods have shown competitive performance on tracking benchmarks in recent years. Generally, DCF based trackers learn a rigid appearance model of the target. However, this reliance on a single rigid appearance model is insufficient in situations where the target undergoes non-rigid transformations. In this paper, we propose a unified formulation for learning a deformable convolution filter. In our framework, the deformable filter is represented as a linear combination of sub-filters. Both the sub-filter coefficients and their relative locations are inferred jointly in our formulation. Experiments are performed on three challenging tracking benchmarks: OTB- 2015, TempleColor and VOT2016. Our approach improves the baseline method, leading to performance comparable to state-of-the-art.