View Code? Open in Web Editor NEW

This project forked from ustc-vim/vegfru

VegFru: A Domain-Specific Dataset for Fine-grained Visual Categorization

CMake 1.18% Makefile 0.28% HTML 0.08% CSS 0.10% Jupyter Notebook 57.69% C++ 32.89% Shell 0.37% Python 3.81% Cuda 2.54% C 0.05% MATLAB 0.36% M 0.01% Protocol Buffer 0.64%

vegfru's Introduction

VegFru

The repository is for the paper "VegFru: A Domain-Specific Dataset for Fine-grained Visual Categorization".

Download Links: [Paper] [Supplementary Material] [Project Page] [Download VegFru Dataset]

The VegFru dataset is available at here and this code is for the implementation of HybridNet.

./caffe-hybridnet: modified from Caffe (https://github.com/BVLC/caffe)
./hybridnet-dataset: the prototxt defining the models for each dataset (FGVC-Aircrafts)

Usage

cd caffe-hybridnet
sh data/jointaircrafts/model/v4_hyvggcpnet/ft_last_layer.sh
sh data/jointaircrafts/model/v4_hyvggcpnet/ft_all.sh

The pretrained models can be downloaded at here.

Citation

Please cite the following paper if you find this useful in your research:

@InProceedings{Hou2017VegFru,
  Title                    = {VegFru: A Domain-Specific Dataset for Fine-grained Visual Categorization},
  Author                   = {Saihui Hou, Yushan Feng and Zilei Wang},
  Booktitle                = {IEEE International Conference on Computer Vision},
  Year                     = {2017}
}

Recommend Projects