Search Multilayer Perceptron-Based Fusion for Efficient and Accurate Siamese Tracking

TL;DR

This paper introduces a novel MLP-based fusion module for Siamese trackers that improves pixel-level interaction efficiency, balancing accuracy and computational cost through neural architecture search, and achieves state-of-the-art results on multiple benchmarks.

Contribution

It proposes a simple MLP-based fusion module with a differentiable architecture search strategy to optimize channel width and depth, enhancing tracker efficiency and accuracy.

Findings

Achieves state-of-the-art accuracy-efficiency trade-offs.

Ranks among top on multiple tracking benchmarks.

Maintains real-time performance on resource-constrained hardware.

Abstract

Siamese visual trackers have recently advanced through increasingly sophisticated fusion mechanisms built on convolutional or Transformer architectures. However, both struggle to deliver pixel-level interactions efficiently on resource-constrained hardware, leading to a persistent accuracy-efficiency imbalance. Motivated by this limitation, we redesign the Siamese neck with a simple yet effective Multilayer Perception (MLP)-based fusion module that enables pixel-level interaction with minimal structural overhead. Nevertheless, naively stacking MLP blocks introduces a new challenge: computational cost can scale quadratically with channel width. To overcome this, we construct a hierarchical search space of carefully designed MLP modules and introduce a customized relaxation strategy that enables differentiable neural architecture search (DNAS) to decouple channel-width optimization from other architectural choices. This targeted decoupling automatically balances channel width and depth, yielding a low-complexity architecture. The resulting tracker achieves state-of-the-art accuracy-efficiency trade-offs. It ranks among the top performers on four general-purpose and three aerial tracking benchmarks, while maintaining real-time performance on both resource-constrained Graphics Processing Units (GPUs) and Neural Processing Units (NPUs).