AttZoom: Attention Zoom for Better Visual Features

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: DeAlcala, Daniel, Morales, Aythami, Fierrez, Julian, Tolosana, Ruben
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916882744868864
author DeAlcala, Daniel
Morales, Aythami
Fierrez, Julian
Tolosana, Ruben
author_facet DeAlcala, Daniel
Morales, Aythami
Fierrez, Julian
Tolosana, Ruben
contents We present Attention Zoom, a modular and model-agnostic spatial attention mechanism designed to improve feature extraction in convolutional neural networks (CNNs). Unlike traditional attention approaches that require architecture-specific integration, our method introduces a standalone layer that spatially emphasizes high-importance regions in the input. We evaluated Attention Zoom on multiple CNN backbones using CIFAR-100 and TinyImageNet, showing consistent improvements in Top-1 and Top-5 classification accuracy. Visual analyses using Grad-CAM and spatial warping reveal that our method encourages fine-grained and diverse attention patterns. Our results confirm the effectiveness and generality of the proposed layer for improving CCNs with minimal architectural overhead.
format Preprint
id arxiv_https___arxiv_org_abs_2508_03625
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AttZoom: Attention Zoom for Better Visual Features
DeAlcala, Daniel
Morales, Aythami
Fierrez, Julian
Tolosana, Ruben
Computer Vision and Pattern Recognition
Artificial Intelligence
We present Attention Zoom, a modular and model-agnostic spatial attention mechanism designed to improve feature extraction in convolutional neural networks (CNNs). Unlike traditional attention approaches that require architecture-specific integration, our method introduces a standalone layer that spatially emphasizes high-importance regions in the input. We evaluated Attention Zoom on multiple CNN backbones using CIFAR-100 and TinyImageNet, showing consistent improvements in Top-1 and Top-5 classification accuracy. Visual analyses using Grad-CAM and spatial warping reveal that our method encourages fine-grained and diverse attention patterns. Our results confirm the effectiveness and generality of the proposed layer for improving CCNs with minimal architectural overhead.
title AttZoom: Attention Zoom for Better Visual Features
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2508.03625