轻量级视觉Transformer，让边缘设备也能高效运行图像识别。
MicroViT: A Vision Transformer with Low Complexity Self Attention for Edge Device
- 用分组卷积减少通道冗余，仅处理部分通道实现低复杂度自注意力。
- 在ImageNet-1K上保持高精度，推理速度比MobileViT快3.6倍。
- 适合手机、IoT等资源受限设备部署，能效提升40%。
视觉变换器（ViT）在多种计算机视觉任务中表现优异，但其高计算需求使其难以在资源有限的边缘设备上应用。本文提出MicroViT，一种专为边缘设备优化的轻量级视觉变换器架构，通过显著降低计算复杂度，在保持高准确率的同时实现高效运行。其核心是高效的单头注意力机制（ESHA），利用分组卷积减少特征冗余，并仅处理部分通道，从而减轻自注意力机制负担。MicroViT采用多阶段MetaFormer架构，堆叠多个MicroViT编码器以提升效率与性能。在ImageNet-1K和COCO数据集上的全面实验表明，MicroViT在保持竞争力准确率的同时，推理速度比MobileViT系列快3.6倍，能耗降低，能效提升40%，适用于移动和边缘设备等资源受限环境。
原文摘要 · Abstract (English)
The Vision Transformer (ViT) has demonstrated state-of-the-art performance in various computer vision tasks, but its high computational demands make it impractical for edge devices with limited resources. This paper presents MicroViT, a lightweight Vision Transformer architecture optimized for edge devices by significantly reducing computational complexity while maintaining high accuracy. The core of MicroViT is the Efficient Single Head Attention (ESHA) mechanism, which utilizes group convolution to reduce feature redundancy and processes only a fraction of the channels, thus lowering the burden of the self-attention mechanism. MicroViT is designed using a multi-stage MetaFormer architecture, stacking multiple MicroViT encoders to enhance efficiency and performance. Comprehensive experiments on the ImageNet-1K and COCO datasets demonstrate that MicroViT achieves competitive accuracy while significantly improving 3.6 faster inference speed and reducing energy consumption with 40% higher efficiency than the MobileViT series, making it suitable for deployment in resource-constrained environments such as mobile and edge devices.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。