2023.4.11 pytorch读取coco数据集以及读取自己建立的coco数据集

1.pytorch读取coco数据集

import torchvision
from PIL import ImageDraw

coco_dataset = torchvision.datasets.CocoDetection(root=r"F:\DATA_bage\val2017",
                                                  annFile=r"F:\DATA_bage\annotations\instances_val2017.json")
image,info = coco_dataset[0]
image_handler = ImageDraw.ImageDraw(image)

for annotation in info:
    x_min,y_min,width,height = annotation['bbox']
    image_handler.rectangle(((x_min,y_min),(x_min+width,y_min+height)))

image.show()

2.pytorch读取自己建立的coco数据集

import torchvision
from PIL import ImageDraw

coco_dataset = torchvision.datasets.CocoDetection(root=r"C:\Users\Administrator\Desktop\code_study\picture",
                                                  annFile=r"C:\you coco direction.json")
image,info = coco_dataset[0]  #取第一张图片
image_handler = ImageDraw.ImageDraw(image)

for annotation in info:
    x_min,y_min,width,height = annotation['bbox']
    image_handler.rectangle(((x_min,y_min),(x_min+width,y_min+height)))

image.show()