https://github.com/pris-cv/fine-grained-visual-analysis-library

FGVCLib is an open-source and well documented library for Fine-grained Visual Classification.
https://github.com/pris-cv/fine-grained-visual-analysis-library

classification computer-vision fine-grained fine-grained-classification fine-grained-visual-categorization

Last synced: about 1 year ago
JSON representation

FGVCLib is an open-source and well documented library for Fine-grained Visual Classification.

• Host: GitHub
• URL: https://github.com/pris-cv/fine-grained-visual-analysis-library
• Owner: PRIS-CV
• License: mit
• Created: 2022-07-02T22:24:56.000Z (about 4 years ago)
• Default Branch: main
• Last Pushed: 2023-12-09T06:59:30.000Z (over 2 years ago)
• Last Synced: 2025-07-07T01:02:49.209Z (about 1 year ago)
• Topics: classification, computer-vision, fine-grained, fine-grained-classification, fine-grained-visual-categorization
• Language: Python
• Homepage:
• Size: 307 KB
• Stars: 38
• Watchers: 3
• Forks: 3
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Fine-grained-Visual-Analysis-Library

## Introduction

FGVCLib is an open-source and well documented library for Fine-grained Visual Classification. It is based on Pytorch with performance and friendly API. Our code is pythonic, and the design is consistent with torchvision. You can easily develop new algorithms, or readily apply existing algorithms.

The branch works with **torch 1.12.1**, **torchvision 0.13.1**.

For more details and the tutorials about the FGVCLib, see [FGVCLib](https://pris-cv-fgvclib.readthedocs.io/en/latest/index.html)

Major features

- **Modular Design**

  We decompose the detection framework into different components and one can easily construct a customized object detection framework by combining different modules.

- **State of the art**

  We implement state-of-the-art methods by the FGVCLib, [PMG](https://arxiv.org/abs/2003.03836v3), [PMG_V2](https://ieeexplore.ieee.org/abstract/document/9609669), [MCL](https://arxiv.org/abs/2002.04264), [API-Net](https://arxiv.org/abs/2002.10191), [CAL](https://ieeexplore.ieee.org/document/9710619), [TransFG](https://ieeexplore.ieee.org/document/9710619), [PIM](https://arxiv.org/abs/2202.03822). 

## Installation

Please refer to [Installation](https://pris-cv-fgvclib.readthedocs.io/en/latest/get_started.html) for installation instructions.

## Getting started 

Please see [get_started.md](https://pris-cv-fgvclib.readthedocs.io/en/latest/get_started.html) for the basic usage of FGVCLib. We provide the tutorials for:

- [with existing data existing model](https://pris-cv-fgvclib.readthedocs.io/en/latest/1_exist_data_model.html)

- [with existing data new model](https://pris-cv-fgvclib.readthedocs.io/en/latest/2_exist_data_new_model.html)

- [learn about apis](https://pris-cv-fgvclib.readthedocs.io/en/latest/tutorials/tutorial1_apis.html)

- [learn about configs](https://pris-cv-fgvclib.readthedocs.io/en/latest/tutorials/tutorial2_configs.html)

- [learn about criterions](https://pris-cv-fgvclib.readthedocs.io/en/latest/tutorials/tutorial3_criterions.html)

- [learn about datasets](https://pris-cv-fgvclib.readthedocs.io/en/latest/tutorials/tutorial4_datasets.html)

- [learn about metrics](https://pris-cv-fgvclib.readthedocs.io/en/latest/tutorials/tutorial5_metrics.html)

- [learn about model](https://pris-cv-fgvclib.readthedocs.io/en/latest/tutorials/tutorial6_model.html)

- [learn about transforms](https://pris-cv-fgvclib.readthedocs.io/en/latest/tutorials/tutorial7_transform.html)

- [learn about the tools](https://pris-cv-fgvclib.readthedocs.io/en/latest/useful_tools.html)

## Overview of Benchmark and Model Zoo



  Architectures



  

    

      

        Fine-grained Visual Classification

      

      

        Other

      

    

    

      

        


            
Baseline_ResNet50

            
Mutual-Channel-Loss

            
PMG-ResNet50

            
PMG_V2_ResNet50

            
API-Net

            
CAL

            
TransFG

            
PIM

      
  

      

      

        


            
visualization

      
  

      

    

  




  Components



  

    

      

        Backbones

      

      

        Encoders

      

      

        Heads

      

      

        Necks

      

      

        Sotas

      

    

    

      

        


            
Resnet

            
VGG

      
  

      

      

        


            
Global Max Pooling

            
Global Avg Pooling

            
Max Pooling 2d

      
  

      

      

        


            
Classifier_1_FC

            
Classifier_2_FC

      
  

      

      

        


            
Multi-scale Convolution neck

      
  

      

      

        


            
Baseline_ResNet50

            
Mutual-Channel-Loss

            
PMG-ResNet50

            
PMG_V2_ResNet50

            
API-Net

            
CAL

            
TransFG

            
PIM

      
  

      

    

  


## Contact

Thanks for your attention! If you have any suggestion or question, you can leave a message here or contact us directly:

- changdongliang@bupt.edu.cn

- mazhanyu@bupt.edu.cn

## Others 

Based on the fgvclib, we have developed an FGVC WeChat applet for fine-grained visual classification in practice, which can be accessed by searching "细粒度图像分类" in WeChat, and there is a demo: https://reurl.cc/rRZE7O.

## Citation

If you find this library useful in your research, please consider citing:

```

@misc{Chang2023,

  author = {Dongliang, Chang and Ruoyi, Du and Xinran, Wang and Yuqi, Yang and Yi-Zhe, Song and Zhanyu, Ma},

  title = {Fine-grained Visual Analysis Library},

  year = {2023},

  publisher = {GitHub},

  journal = {GitHub repository},

  howpublished = {\url{https://github.com/PRIS-CV/Fine-grained-Visual-Analysis-Library}}

}

```