提出高效自回归追踪框架，兼顾高速与高精度。
FARTrack: Fast Autoregressive Visual Tracking with High Performance
- 采用任务特异性自蒸馏与帧间稀疏化提升推理效率
- 在GOT-10k上实现70.6% AO与343 FPS（GPU）
- 适合部署于资源受限设备的实时视觉追踪
推理速度与跟踪性能是视觉追踪领域的两个关键评价指标。然而，高性能追踪器通常存在处理速度慢的问题，难以在资源受限设备上部署。为此，我们提出FARTrack，一种快速自回归追踪框架。由于自回归强调轨迹序列的时间特性，可在各类设备上实现高效执行的同时保持高性能。FARTrack引入任务特异性自蒸馏与帧间自回归稀疏化，分别从浅层精准蒸馏与冗余到关键标记优化的角度设计。任务特异性自蒸馏通过逐层提取任务特定标记实现模型压缩，在不依赖人工指定师生层配对的情况下提升推理速度；帧间自回归稀疏化则按序压缩多个模板，在学习时间全局最优稀疏策略的同时避免额外运行开销。FARTrack展现出卓越的速度与竞争力表现：在GOT-10k上实现实时70.6% AO；最快模型在GPU上达到343 FPS，CPU上达121 FPS。
原文摘要 · Abstract (English)
Inference speed and tracking performance are two critical evaluation metrics in the field of visual tracking. However, high-performance trackers often suffer from slow processing speeds, making them impractical for deployment on resource-constrained devices. To alleviate this issue, we propose FARTrack, a Fast Auto-Regressive Tracking framework. Since autoregression emphasizes the temporal nature of the trajectory sequence, it can maintain high performance while achieving efficient execution across various devices. FARTrack introduces Task-Specific Self-Distillation and Inter-frame Autoregressive Sparsification, designed from the perspectives of shallow-yet-accurate distillation and redundant-to-essential token optimization, respectively. Task-Specific Self-Distillation achieves model compression by distilling task-specific tokens layer by layer, enhancing the model's inference speed while avoiding suboptimal manual teacher-student layer pairs assignments. Meanwhile, Inter-frame Autoregressive Sparsification sequentially condenses multiple templates, avoiding additional runtime overhead while learning a temporally-global optimal sparsification strategy. FARTrack demonstrates outstanding speed and competitive performance. It delivers an AO of 70.6% on GOT-10k in real-time. Beyond, our fastest model achieves a speed of 343 FPS on the GPU and 121 FPS on the CPU.
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。