arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2312.00596cs.CVcs.AI
收藏

BCN:用于图像分类的批通道归一化

BCN: Batch Channel Normalization for Image Classification

Huazhong University of Science and Technology(华中科技大学)

Afifa Khaled, Chao Li, Jia Ning, Kun He

AI总结:

本文提出批通道归一化(BCN),通过自适应结合批归一化和层归一化的优势,可无缝集成到 CNN 和 Vision Transformer 中,提升图像分类性能。

摘要:

归一化技术在深度学习领域得到了广泛应用,因为它们能够支持更高的学习率,并且对初始化的要求不那么严格。然而,流行归一化技术的有效性通常仅限于特定领域。与标准的批归一化(BN)和层归一化(LN)不同——BN 沿 (N,H,W) 维度计算均值和方差,LN 沿 (C,H,W) 维度计算均值和方差(N、C、H 和 W 分别表示批、通道、空间高度和宽度维度)——本文提出了一种名为批通道归一化(BCN)的新型归一化技术。为了同时利用通道和批依赖性,并根据特定数据集或任务自适应地结合 BN 和 LN 的优势,BCN 分别沿 (N, H, W) 和 (C, H, W) 轴对输入进行归一化,然后基于自适应参数将归一化后的输出进行组合。作为一个基本模块,BCN 可以轻松集成到现有模型中,用于计算机视觉领域的各种应用。实验结果表明,所提出的技术可以无缝应用于各种版本的 CNN 或 Vision Transformer 架构。代码已公开,地址为 https://github.com/AfifaKhaled/BatchChannel-Normalization。

英文摘要:

Normalization techniques have been widely used in the field of deep learning due to their capability of enabling higher learning rates and are less careful in initialization. However, the effectiveness of popular normalization technologies is typically limited to specific areas. Unlike the standard Batch Normalization (BN) and Layer Normalization (LN), where BN computes the mean and variance along the (N,H,W) dimensions and LN computes the mean and variance along the (C,H,W) dimensions (N, C, H and W are the batch, channel, spatial height and width dimension, respectively), this paper presents a novel normalization technique called Batch Channel Normalization (BCN). To exploit both the channel and batch dependence and adaptively and combine the advantages of BN and LN based on specific datasets or tasks, BCN separately normalizes inputs along the (N, H, W) and (C, H, W) axes, then combines the normalized outputs based on adaptive parameters. As a basic block, BCN can be easily integrated into existing models for various applications in the field of computer vision. Empirical results show that the proposed technique can be seamlessly applied to various versions of CNN or Vision Transformer architecture. The code is publicly available at https://github.com/AfifaKhaled/BatchChannel-Normalization

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑