提出自适应视觉变换器，实时高效追踪无人机，视角不变性强。
Learning an Adaptive and View-Invariant Vision Transformer for Real-Time UAV Tracking
- 动态激活注意力模块，按需计算提升效率
- 通过互信息最大化学习视角不变特征，抗极端视角变化
- 多教师知识蒸馏增强模型泛化性，适合资源受限设备
基于Transformer的模型虽提升了视觉追踪性能，但在资源受限设备上仍难实现实时运行，尤其在无人机追踪场景。为此，我们提出AVTrack，一种自适应计算追踪框架，通过激活模块（AM）动态选择性启用视觉变换器块，实现对ViT结构的动态优化。为应对极端视角变化，提出通过互信息（MI）最大化学习视角不变表示。此外，提出AVTrack-MD，一种结合新型基于互信息最大化的多教师知识蒸馏框架。利用多个现成的AVTrack模型作为教师，最大化其聚合软特征与学生模型对应软特征之间的互信息，提升学生模型的泛化能力与性能，尤其在噪声环境下表现更优。大量实验表明，AVTrack-MD在性能接近原版的同时，模型复杂度降低，平均追踪速度提升超过17%。代码已公开：https://github.com/wuyou3474/AVTrack。
原文摘要 · Abstract (English)
Transformer-based models have improved visual tracking, but most still cannot run in real time on resource-limited devices, especially for unmanned aerial vehicle (UAV) tracking. To achieve a better balance between performance and efficiency, we propose AVTrack, an adaptive computation tracking framework that adaptively activates transformer blocks through an Activation Module (AM), which dynamically optimizes the ViT architecture by selectively engaging relevant components. To address extreme viewpoint variations, we propose to learn view-invariant representations via mutual information (MI) maximization. In addition, we propose AVTrack-MD, an enhanced tracker incorporating a novel MI maximization-based multi-teacher knowledge distillation framework. Leveraging multiple off-the-shelf AVTrack models as teachers, we maximize the MI between their aggregated softened features and the corresponding softened feature of the student model, improving the generalization and performance of the student, especially under noisy conditions. Extensive experiments show that AVTrack-MD achieves performance comparable to AVTrack's performance while reducing model complexity and boosting average tracking speed by over 17\%. Codes is available at: https://github.com/wuyou3474/AVTrack.
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。