Enregistré dans:
Détails bibliographiques
Auteurs principaux: Kriuk, Boris, Gill, Simranjit Kaur, Aslam, Shoaib, Fakhrutdinov, Amir
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2504.09852
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915240854159360
author Kriuk, Boris
Gill, Simranjit Kaur
Aslam, Shoaib
Fakhrutdinov, Amir
author_facet Kriuk, Boris
Gill, Simranjit Kaur
Aslam, Shoaib
Fakhrutdinov, Amir
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2504_09852
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GFT: Gradient Focal Transformer
Kriuk, Boris
Gill, Simranjit Kaur
Aslam, Shoaib
Fakhrutdinov, Amir
Computer Vision and Pattern Recognition
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.
title GFT: Gradient Focal Transformer
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.09852