AttZoom: Attention Zoom for Better Visual Features

2025/08/05 by DeAlcala, Daniel, Morales, Aythami, Fierrez, Julian +1 · 1 citation
#Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2508.03625

Abstract

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.

Citations

• PBa-LLM: Privacy- and Bias-aware NLP using Named-Entity Recognition (NER)
• Are Vision-Language Models Ready for Dietary Assessment? Exploring the Next Frontier in AI-Powered Food Image Recognition
• Brain-inspired Artificial Intelligence: A Comprehensive Review
• VAAD: Visual Attention Analysis Dashboard applied to e-Learning
• Visual Attention Methods in Deep Learning: An In-Depth Survey
• A General Survey on Attention Mechanisms in Deep Learning
• A ConvNet for the 2020s
• A ConvNet for the 2020s
• Attention mechanisms in computer vision: A survey
• FcaNet: Frequency Channel Attention Networks
• Fusing CNNs and statistical indicators to improve image classification
• Deformable DETR: Deformable Transformers for End-to-End Object Detection
• Rethinking Attention with Performers
• End-to-End Object Detection with Transformers
• ResNeSt: Split-Attention Networks
• One-Shot Object Detection with Co-Attention and Co-Excitation
• ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks
• ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks
• Gated Channel Transformation for Visual Recognition
• Optuna: A Next-generation Hyperparameter Optimization Framework
• Pyramid Feature Attention Network for Saliency detection
• SensitiveNets: Learning Agnostic Representations with Application to Face Images
• Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
• CBAM: Convolutional Block Attention Module
• BAM: Bottleneck Attention Module
• Squeeze-and-Excitation Networks
• A Downsampled Variant of ImageNet as an Alternative to the CIFAR datasets
• Attention Is All You Need
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Aggregated Residual Transformations for Deep Neural Networks
• Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
• Densely Connected Convolutional Networks
• Deep Residual Learning for Image Recognition
• DRAW: A Recurrent Neural Network For Image Generation
• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• Recurrent Models of Visual Attention
• Long Short-Term Memory
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
• BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Cited by

• Privacy-Aware Detection of Fake Identity Documents: Methodology, Benchmark, and Improved Algorithms (FakeIDet2)

Related

• STEAM: Squeeze and Transform Enhanced Attention Module
• CNNtention: Can CNNs do better with Attention?
• Interpreting and Improving Attention From the Perspective of Large Kernel Convolution
• TDAM: Top-Down Attention Module for Contextually Guided Feature Selection in CNNs
• Feature-based Attention in Convolutional Neural Networks