Search papers, labs, and topics across Lattice.
This paper introduces CrossMambaTuning, a framework that enhances the adaptation of pretrained learned image compression models for machine vision tasks by integrating State Space Models with cross-layer interaction mechanisms. The approach employs a Mamba adapter with task-specific prompts and multi-scale branching to effectively capture both local features and global dependencies, while the Scale-Invariant Cross-Layer Adapter (SICA) reduces redundancy through parameter sharing. Experimental results show that CrossMambaTuning achieves state-of-the-art performance across various tasks, achieving a 72% reduction in parameter overhead compared to existing methods.
Achieving state-of-the-art performance with 72% fewer parameters, CrossMambaTuning redefines efficiency in adapting image compression models for machine vision.
To reduce deployment cost and retraining overhead, adapting pretrained learned image compression (LIC) models to downstream machine vision tasks has attracted growing attention. However, existing methods typically insert fine-tuning modules independently into frozen backbones, lacking explicit mechanisms for cross-layer coordination. To address this limitation, we propose a novel framework named CrossMambaTuning, which integrates State Space Models with cross-layer interaction mechanisms for parameter-efficient fine-tuning. Specifically, we design an efficient Mamba adapter equipped with task-specific prompts and multi-scale branching to precisely capture both local features and global dependencies. Furthermore, we introduce a Scale-Invariant Cross-Layer Adapter (SICA) utilizing a parameter-sharing strategy to fuse task information across different scales and reduce redundancy. Extensive experiments demonstrate that CrossMambaTuning achieves state-of-the-art (SOTA) performance on multiple machine vision tasks, reducing parameter overhead by 72\% compared to SOTA methods. Code is available at https://github.com/rsr1123/CrossMambaTuning.
