提出一种矩形注意力模块，让模型聚焦图像关键区域并提升泛化能力。
Convolutional Rectangular Attention Module
- 用5个参数控制矩形注意力区域，避免传统方法的不规则边界。
- 在多个数据集上性能超越传统逐位置注意力，提升模型准确率。
- 适合需要可解释性的视觉任务，如医学影像分析或目标检测。
本文提出一种新型空间注意力模块，可无缝集成至任意卷积网络中。该模块引导模型关注图像中最具判别性的部分，通过端到端训练提升性能。传统方法通常以逐位置方式生成空间注意力图，导致边界不规则，影响模型对新样本的泛化能力。本文方法将注意力区域约束为矩形，仅用5个参数进行参数化，从而提升稳定性和泛化性。实验表明，本方法系统优于传统的逐位置注意力机制，为卷积模型提供了一种新颖且有效的空间注意力机制。此外，该模块还增强了可解释性，能明确指示模型在预测时关注输入图像的哪一部分。
原文摘要 · Abstract (English)
In this paper, we introduce a novel spatial attention module that can be easily integrated to any convolutional network. This module guides the model to pay attention to the most discriminative part of an image. This enables the model to attain a better performance by an end-to-end training. In conventional approaches, a spatial attention map is typically generated in a position-wise manner. Thus, it is often resulting in irregular boundaries and so can hamper generalization to new samples. In our method, the attention region is constrained to be rectangular. This rectangle is parametrized by only 5 parameters, allowing for a better stability and generalization to new samples. In our experiments, our method systematically outperforms the position-wise counterpart. So that, we provide a novel useful spatial attention mechanism for convolutional models. Besides, our module also provides the interpretability regarding the \textit{where to look} question, as it helps to know the part of the input on which the model focuses to produce the prediction.
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。