通过动态裁剪冗余层，让ViT在无人机追踪中更快更准。
Similarity-Guided Layer-Adaptive Vision Transformer for UAV Tracking
- 基于相似性判断，自动关闭重复的网络层。
- 在6个数据集上实现实时速度与高精度平衡。
- 适合需要轻量化实时追踪的无人机场景。
视觉变压器（ViTs）已成为视觉追踪的主流主干网络，但完整结构在强调效率的无人机追踪任务中过于臃肿。本研究发现，轻量级ViT追踪器中的许多层会学习到高度冗余且重复的目标表征。基于此，提出一种相似性引导的层自适应方法，动态禁用大量表征相似的层，仅保留其中最优的一层，以实现更好的精度-速度权衡。将该方法融入现有ViTs后，构建出高效、可部署的相似性引导层自适应框架SGLATrack，适用于实时无人机追踪。在六个追踪基准上的大量实验验证了其有效性，结果表明SGLATrack在保持竞争性追踪精度的同时，实现了最先进的实时速度。代码与模型见https://github.com/GXNU-ZhongLab/SGLATrack。
原文摘要 · Abstract (English)
Vision transformers (ViTs) have emerged as a popular backbone for visual tracking. However, complete ViT architectures are too cumbersome to deploy for unmanned aerial vehicle (UAV) tracking which extremely emphasizes efficiency. In this study, we discover that many layers within lightweight ViT-based trackers tend to learn relatively redundant and repetitive target representations. Based on this observation, we propose a similarity-guided layer adaptation approach to optimize the structure of ViTs. Our approach dynamically disables a large number of representation-similar layers and selectively retains only a single optimal layer among them, aiming to achieve a better accuracy-speed trade-off. By incorporating this approach into existing ViTs, we tailor previously complete ViT architectures into an efficient similarity-guided layer-adaptive framework, namely SGLATrack, for real-time UAV tracking. Extensive experiments on six tracking benchmarks verify the effectiveness of the proposed approach, and show that our SGLATrack achieves a state-of-the-art real-time speed while maintaining competitive tracking precision. Codes and models are available at https://github.com/GXNU-ZhongLab/SGLATrack.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。