图像标注及分类数据集

xuqian

图像标注及分类数据集：

1.数据名称：INRIA features for image annotation and classification data sets
  主要包含：corel5k, iaprtc12, espgame, pascal07, mirflickr 共5个数据集

2.数据内容：
  (1) A dictionary: the list of words used for annotating the images.
  (2) Lists of files of train and test sets. You will have to figure out the correspondence of these lists with the set of images you collected, but this shouldn't be too hard.
  (3) Annotation files for the train/test sets: matrix containing binary values to encode the annotations. Columns correspond to words in the dictionary, rows correspond to images as in the lists
  (4) 15 Descriptor files for the train/test sets: matrices containing visual descriptors for the images: Gist, DenseSift, DenseSiftV3H1, HarrisSift, HarrisSiftV3H1, DenseHue, DenseHueV3H1, HarrisHue, HarrisHueV3H1, Rgb, RgbV3H1, Lab, LabV3H1, Hsv, HsvV3H1

3.数据下载：
  链接：http://pan.baidu.com/s/1geQ1pv1
  密码：w13x

4.数据来源：http://lear.inrialpes.fr/people/guillaumin/data.php

5.数据读取：
  调用”vectools_matlab”文件夹下的vec_read.m和vec_write.m函数
  下面给出一个样例：
  %% load the train and test data, Espgame, Dense sift features
  dataset_train=double(vec_read('espgame_train_DenseSift.hvecs'));
  dataset_test=double(vec_read('espgame_test_DenseSift.hvecs'));

zhuwencheng

下载的数据放到 D:\ 目录下（如果放在其他的目录下，修改文件位置即可），处理方式：
dataset = 'pascal07';
fea = 'DenseSift';

switch dataset
    case 'espgame'
        tr_d = double(vec_read(['D:\ImageMulti-labelData\espgame.20091111\espgame_train_' fea '.hvecs']));
        tt_d = double(vec_read(['D:\ImageMulti-labelData\espgame.20091111\espgame_test_' fea '.hvecs']));
        tt_l = double(vec_read('D:\ImageMulti-labelData\espgame.20091111\espgame_test_annot.hvecs'));
        tr_l = double(vec_read('D:\ImageMulti-labelData\espgame.20091111\espgame_train_annot.hvecs'));
    case 'iaprtc12'
        tr_d = double(vec_read(['D:\ImageMulti-labelData\iaprtc12.20091111\iaprtc12_train_' fea '.hvecs']));
        tt_d = double(vec_read(['D:\ImageMulti-labelData\iaprtc12.20091111\iaprtc12_test_' fea '.hvecs']));
        tt_l = double(vec_read('D:\ImageMulti-labelData\iaprtc12.20091111\iaprtc12_test_annot.hvecs'));
        tr_l = double(vec_read('D:\ImageMulti-labelData\iaprtc12.20091111\iaprtc12_train_annot.hvecs'));
    case 'mirflickr'
        tr_d = double(vec_read(['D:\ImageMulti-labelData\mirflickr.20101118\mirflickr_train_' fea '.hvecs']));
        tt_d = double(vec_read(['D:\ImageMulti-labelData\mirflickr.20101118\mirflickr_test_' fea '.hvecs']));

        tt_l = double(load('D:\ImageMulti-labelData\mirflickr.20101118\mirflickr_test_classes.txt'));
        tr_l = double(load('D:\ImageMulti-labelData\mirflickr.20101118\mirflickr_train_classes.txt'));
    case 'pascal07'
        tr_d = double(vec_read(['D:\ImageMulti-labelData\pascal07.20100609\pascal07_train_' fea '.hvecs']));
        tt_d = double(vec_read(['D:\ImageMulti-labelData\pascal07.20100609\pascal07_test_' fea '.hvecs']));

        tt_l = double(load('D:\ImageMulti-labelData\pascal07.20100609\pascal07_test_classes.txt'));
        tr_l = double(load('D:\ImageMulti-labelData\pascal07.20100609\pascal07_train_classes.txt'));
end