Deformable Siamese Attention Networks for Visual Object Tracking

Siamese-based trackers have achieved excellent performance on visual object tracking. However, the target template is not updated online, and the features of the target template and search image are computed independently in a Siamese architecture. In this paper, we propose Deformable Siamese Attention Networks, referred to as Sia-mAttn, by introducing a new Siamese attention mechanism that computes deformable self-attention and crossattention. The self-attention learns strong context information via spatial attention, and selectively emphasizes interdependent channel-wise features with channel attention. The cross-attention is capable of aggregating rich contextual interdependencies between the target template and the search image, providing an implicit manner to adaptively update the target template. In addition, we design a region refinement module that computes depth-wise cross correlations between the attentional features for more accurate tracking. We conduct experiments on six benchmarks, where our method achieves new state-of-the-art results, outperforming the strong baseline, SiamRPN++ [24], by 0.464→0.537 and 0.415→0.470 EAO on VOT 2016 and 2018. Our code is available at: https://github.com/msight- tech/research-siamattn.