用Fiftyone下载OpenImage Datasetv6数据集

参考链接：https://voxel51.com/docs/fiftyone/tutorials/open_images.html

一、Openimage

Openimage大约有九万张图片，包含六千多种类别，包括图像检测、图像分割等，在下载之前可以在explore栏先搜索自己想要的类别
先选Type再搜！第一次在搜索glove的时候没有修改type，选择的是segmentation，没有找到

二、安装fiftyone下载器

pip install fiftyone
!pip install tensorflow torch torchvision umap-learn
!pip install ipywidgets>=7.5

下载demo测试

import fiftyone as fo
import fiftyone.zoo as foz
dataset = foz.load_zoo_dataset(
    "open-images-v6",
    split="validation",
    max_samples=100,
    seed=51,
    shuffle=True,
)

默认的下载路径在C盘用户文件夹下的"fiftyone"文件里，注意不是".fiftyone"哦


下好了！
下载特定的数据集，数据集的数量以及下载路径
参数说明：
label_types- a list of label types to load. The supported values are (“detections”, “classifications”, “segmentations”, “relationships”). By default, all available labels types will be loaded. Specifying [] will load only the images
classes- a list of classes of interest. If specified, only samples with at least one object, segmentation, or image-level label in the specified classes will be downloaded
attrs- a list of attributes of interest. If specified, only download samples if they contain at least one attribute in attrs or one class in classes (only applicable when label_types contains “relationships”)
load_hierarchy- whether to load the class hierarchy into dataset.info[“hierarchy”]
image_ids- an array of specific image IDs to download
image_ids_file - a path to a .txt, .csv, or .json file containing image IDs to download
In addition, like all other zoo datasets, you can specify:
max_samples- the maximum number of samples to load
shuffle- whether to randomly chose which samples to load if max_samples is given
seed- a random seed to use when shuffling

代码

if __name__ == '__main__':
    dataset_test = foz.load_zoo_dataset(
        "open-images-v6",
        split="train",# 指定下载数据集
        classes=["Helmet","Glove"],
        shuffle=True,
         max_samples=100, # 指定下载图片数
        only_matching=True,  
        label_types=["detections"],  # 指定下载目标检测的类型,detections,segmentation，relationships，classifications
        dataset_dir="D:\\RAS\OpenImg\Helmet_Glove1",# 保存的路径
        num_workers=2,  # 指定工作进程数
    )

数据格式转换

转换csv格式变成yolov5标准格式

···
import fiftyone as fo
import fiftyone.zoo as foz
import fiftyone.zoo as foz
from fiftyone import ViewField as F

dataset = foz.load_zoo_dataset(“open-images-v6”,
split=“train”,
max_samples=10,
dataset_dir=“E:\Helmet_Glove\Fiftyone”,
label_types=[‘detections’],
classes=[‘Helmet’],
seed=51,
shuffle=True,)

dataset = dataset.filter_labels(“detections”, F(“label”).is_in([“Helmet”]))
dataset.export(
export_dir=‘E:\Helmet_Glove\Fiftyone\datasets’, # 输出路径
dataset_type=fo.types.YOLOv5Dataset,
label_field=‘detections’,
)
···