Two-stream Beats One-stream: Asymmetric Siamese Network for Efficient Visual Tracking
This work addresses the need for efficient tracking on edge devices, offering an incremental improvement over existing lightweight trackers.
The paper tackles the problem of efficient visual tracking for resource-constrained platforms by proposing an asymmetric Siamese network that disentangles template and search streams to reduce redundancy. It achieves superior speed-precision trade-offs, with AsymTrack-T reaching 60.8% AUC on LaSOT and up to 224 FPS on GPU, surpassing prior methods like HiT-Tiny by 6.0% AUC.
Efficient tracking has garnered attention for its ability to operate on resource-constrained platforms for real-world deployment beyond desktop GPUs. Current efficient trackers mainly follow precision-oriented trackers, adopting a one-stream framework with lightweight modules. However, blindly adhering to the one-stream paradigm may not be optimal, as incorporating template computation in every frame leads to redundancy, and pervasive semantic interaction between template and search region places stress on edge devices. In this work, we propose a novel asymmetric Siamese tracker named \textbf{AsymTrack} for efficient tracking. AsymTrack disentangles template and search streams into separate branches, with template computing only once during initialization to generate modulation signals. Building on this architecture, we devise an efficient template modulation mechanism to unidirectional inject crucial cues into the search features, and design an object perception enhancement module that integrates abstract semantics and local details to overcome the limited representation in lightweight tracker. Extensive experiments demonstrate that AsymTrack offers superior speed-precision trade-offs across different platforms compared to the current state-of-the-arts. For instance, AsymTrack-T achieves 60.8\% AUC on LaSOT and 224/81/84 FPS on GPU/CPU/AGX, surpassing HiT-Tiny by 6.0\% AUC with higher speeds. The code is available at https://github.com/jiawen-zhu/AsymTrack.