GFT通过梯度聚焦机制，让模型自动关注细粒度图像中的关键区域，提升识别精度与效率。
GFT: Gradient Focal Transformer
- 基于梯度流分析动态聚焦判别性特征，自适应选择关键区域。
- 在飞机、食物和COCO数据集上达到顶尖准确率，参数量93M。
- 适合需要高精度与可解释性的实际部署场景，如工业质检与医疗影像。
细粒度图像分类（FGIC）是计算机视觉中的一项复杂任务，要求模型区分具有细微局部差异的类别。传统基于CNN的模型虽擅长局部特征提取，但难以捕捉全局上下文；而近年基于ViT的模型虽利用注意力机制增强定位能力，却缺乏对真正判别区域的自适应聚焦。已有工作如TransFG引入部件感知的令牌选择以提升注意力定位，但仍存在计算效率低、注意力区域选择灵活性差及复杂环境下细节聚焦不足的问题。本文提出一种新型基于ViT的框架GFT（Gradient Focal Transformer），结合梯度注意力学习对齐（GALA）机制，通过分析注意力梯度流动态优先处理类别判别性特征，并采用渐进式块选择（PPS）策略逐步过滤不相关信息，降低计算开销同时增强对微小细节的敏感性。GFT在FGVC Aircraft、Food-101和COCO数据集上以9300万参数达到当前最优性能，显著优于现有基于ViT的先进细粒度分类模型，在效率方面也表现更优。通过融合全局上下文与局部细节提取，GFT为细粒度识别设立了新基准，提供适用于真实场景部署的可解释解决方案。
原文摘要 · Abstract (English)
Fine-Grained Image Classification (FGIC) remains a complex task in computer vision, as it requires models to distinguish between categories with subtle localized visual differences. Well-studied CNN-based models, while strong in local feature extraction, often fail to capture the global context required for fine-grained recognition, while more recent ViT-backboned models address FGIC with attention-driven mechanisms but lack the ability to adaptively focus on truly discriminative regions. TransFG and other ViT-based extensions introduced part-aware token selection to enhance attention localization, yet they still struggle with computational efficiency, attention region selection flexibility, and detail-focus narrative in complex environments. This paper introduces GFT (Gradient Focal Transformer), a new ViT-derived framework created for FGIC tasks. GFT integrates the Gradient Attention Learning Alignment (GALA) mechanism to dynamically prioritize class-discriminative features by analyzing attention gradient flow. Coupled with a Progressive Patch Selection (PPS) strategy, the model progressively filters out less informative regions, reducing computational overhead while enhancing sensitivity to fine details. GFT achieves SOTA accuracy on FGVC Aircraft, Food-101, and COCO datasets with 93M parameters, outperforming ViT-based advanced FGIC models in efficiency. By bridging global context and localized detail extraction, GFT sets a new benchmark in fine-grained recognition, offering interpretable solutions for real-world deployment scenarios.
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。