Vision Mamba Distillation for Low-resolution Fine-grained Image Classification

TL;DR

This paper introduces Vision Mamba Distillation (ViMD), a lightweight approach that enhances low-resolution fine-grained image classification by combining a Mamba-based network with multi-level knowledge distillation, achieving state-of-the-art results with fewer resources.

Contribution

The paper proposes a novel lightweight super-resolution Mamba classification network and a multi-level Mamba knowledge distillation method for efficient low-resolution fine-grained image classification.

Findings

ViMD achieves state-of-the-art accuracy on seven datasets.

It outperforms existing methods with fewer parameters and FLOPs.

The approach is suitable for embedded device applications.

Abstract

Low-resolution fine-grained image classification has recently made significant progress, largely thanks to the super-resolution techniques and knowledge distillation methods. However, these approaches lead to an exponential increase in the number of parameters and computational complexity of models. In order to solve this problem, in this letter, we propose a Vision Mamba Distillation (ViMD) approach to enhance the effectiveness and efficiency of low-resolution fine-grained image classification. Concretely, a lightweight super-resolution vision Mamba classification network (SRVM-Net) is proposed to improve its capability for extracting visual features by redesigning the classification sub-network with Mamba modeling. Moreover, we design a novel multi-level Mamba knowledge distillation loss boosting the performance, which can transfer prior knowledge obtained from a High-resolution Vision Mamba classification Network (HRVM-Net) as a teacher into the proposed SRVM-Net as a student. Extensive experiments on seven public fine-grained classification datasets related to benchmarks confirm our ViMD achieves a new state-of-the-art performance. While having higher accuracy, ViMD outperforms similar methods with fewer parameters and FLOPs, which is more suitable for embedded device applications. Code is available at https://github.com/boa2004plaust/ViMD.