Scalable Visual State Space Model with Fractal Scanning

TL;DR

This paper introduces a fractal scanning approach for serializing image patches in State Space Models, significantly improving their ability to model complex spatial patterns and outperforming existing methods in vision tasks.

Contribution

The paper proposes using fractal scanning curves for patch serialization in SSMs, addressing limitations of linear methods and enhancing performance in vision tasks.

Findings

Fractal scanning improves spatial relationship modeling.

Enhanced performance in image classification, detection, and segmentation.

Outperforms existing serialization methods in SSMs.

Abstract

Foundational models have significantly advanced in natural language processing (NLP) and computer vision (CV), with the Transformer architecture becoming a standard backbone. However, the Transformer's quadratic complexity poses challenges for handling longer sequences and higher resolution images. To address this challenge, State Space Models (SSMs) like Mamba have emerged as efficient alternatives, initially matching Transformer performance in NLP tasks and later surpassing Vision Transformers (ViTs) in various CV tasks. To improve the performance of SSMs, one crucial aspect is effective serialization of image patches. Existing methods, relying on linear scanning curves, often fail to capture complex spatial relationships and produce repetitive patterns, leading to biases. To address these limitations, we propose using fractal scanning curves for patch serialization. Fractal curves maintain high spatial proximity and adapt to different image resolutions, avoiding redundancy and enhancing SSMs' ability to model complex patterns accurately. We validate our method in image classification, detection, and segmentation tasks, and the superior performance validates its effectiveness.