Cross-modal Orthogonal High-rank Augmentation for RGB-Event Transformer-trackers

TL;DR

This paper introduces a novel plug-and-play augmentation method using mask modeling and orthogonal high-rank loss to enhance pre-trained vision Transformers for cross-modal RGB and event data object tracking, significantly improving tracking performance.

Contribution

It proposes a new augmentation strategy with mask modeling and orthogonal high-rank regularization to better utilize pre-trained ViTs for cross-modal tracking tasks.

Findings

Significant improvements in tracking precision and success rate.

Effective bridging of modality gaps with simple augmentations.

Enhanced cross-modal interaction in vision Transformers.

Abstract

This paper addresses the problem of cross-modal object tracking from RGB videos and event data. Rather than constructing a complex cross-modal fusion network, we explore the great potential of a pre-trained vision Transformer (ViT). Particularly, we delicately investigate plug-and-play training augmentations that encourage the ViT to bridge the vast distribution gap between the two modalities, enabling comprehensive cross-modal information interaction and thus enhancing its ability. Specifically, we propose a mask modeling strategy that randomly masks a specific modality of some tokens to enforce the interaction between tokens from different modalities interacting proactively. To mitigate network oscillations resulting from the masking strategy and further amplify its positive effect, we then theoretically propose an orthogonal high-rank loss to regularize the attention matrix. Extensive experiments demonstrate that our plug-and-play training augmentation techniques can significantly boost state-of-the-art one-stream and twostream trackers to a large extent in terms of both tracking precision and success rate. Our new perspective and findings will potentially bring insights to the field of leveraging powerful pre-trained ViTs to model cross-modal data. The code will be publicly available.