Foreground-Centric learning improves robustness in fine-grained visual recognition

doi:10.1007/s11760-025-04758-5

2025-09-22

PRE

Ali Alfatemi *

Mohamed Rahouti

Majjed Al-Qatf

Senthil Kumar Jagatheesaperumal

DOI：10.1007/s11760-025-04758-5

原文链接

原文求助

摘要

En 中文

细粒度视觉识别——区分由细微、局部线索差异化的类别——在真实场景中常因背景杂波和姿态变化主导信号而性能下降。我们采用一个简单原则：在识别前隔离主体。通过构建主体中心视角，并在前景而非完整场景上进行训练，决策被导向诊断性形态而非偶然性上下文。在视觉相似物种的挑战性生态案例研究中，该方法在准确率和每类可靠性上实现了持续提升，对易混淆类别改进最为显著。该流程是模块化的、标注轻量且与多样的学习架构和部署环境兼容。我们发布开放资源及可重复性清单以支持透明评估；源代码和支持数据可访问https://github.com/AliAlfatemi/warbler_yolo。除生物多样性监测外，前景中心学习为农业、产品检测和生物医学成像等领域提供了通用的更可靠细粒度识别路径。

Keyword:

Deep learning

Fine-grained classification

Birds

YOLO

Instance segmentation

期刊

Signal Image and Video Processing

IF:

2.1

论文数:

908

被引数:

4.6K

机构

暂无机构信息

引用论文

YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information YOLOv9: 使用可编程梯度信息学习您想要学习的内容

Lecture Notes in Computer Science

IF0

2024-10-26

PREAI

Chien-Yao Wang; I-Hau Yeh; Hong-Yuan Mark Liao

Bird Image Classification using Convolutional Neural Network Transfer Learning Architectures 基于卷积神经网络迁移学习架构的鸟类图像分类

International Journal of Advanced Computer Science and Applications

IF0

2023-01-01

OAAI

Asmita Manna; Nilam Upasani; Shubham Jadhav; Ruturaj Mane; Rutuja Chaudhari; Vishal Chatre

Deep filter banks for texture recognition and segmentation

2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

IF0

2015-06-01

PREAI

Mircea Cimpoi; Subhransu Maji; Andrea Vedaldi

Large Scale Fine-Grained Categorization and Domain-Specific Transfer Learning

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

IF0

2018-06-01

OAAI

Yin Cui; Yang Song; Chen Sun; Andrew Howard; Serge Belongie

Fine-grained recognition without part annotations

2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

IF0

2015-06-01

PREAI

Jonathan Krause; Hailin Jin; Jianchao Yang; Li Fei-Fei

Mask R-CNN 面具r-cnn

2017 IEEE International Conference on Computer Vision (ICCV)

IF0

2017-10-01

PREAI

Kaiming He; Georgia Gkioxari; Piotr Dollar; Ross Girshick

Advancing NCAA March Madness Forecasts Through Deep Learning and Combinatorial Fusion Analysis

IF0

PREAI

Alfatemi,Ali; Rahouti,Mohamed; Hsu,Frank; Schweikert,Christina

The Wood Image Analysis and Dataset (WIAD): Open-access visual analysis tools to advance the ecological data revolution

METHODS IN ECOLOGY AND EVOLUTION

IF6.2

2021-09-30

OAAI

Rademacher, Tim; Seyednasrollah, Bijan; Basler, David J.; Cheng, Jian; Mandra, Tessa; Miller, Elise; Lin, Zuid; Orwig, David A.; Pederson, Neil; Pfister, Hanspeter; Wei, Donglai; Yao, Li; Richardson, Andrew D.

Ecological data sharing

ECOLOGICAL INFORMATICS

IF7.3

2015-09-01

166

OAAI

Michener, William K.

Refining Bird Species Identification through GAN-Enhanced Data Augmentation and Deep Learning Models

Procedia Computer Science

IF0

2024-01-01

PREAI

Alfatemi,Ali; Jamal,Sarah A.L.; Paykari,Nasim; Rahouti,Mohamed; Amin,Ruhul; Chehri,Abdellah

查看更多内容