Loading ADE20K dataset

vision
1

Hi,

I am trying to train PSPNet (Pyramid Scene Parsing Network) for semantic segmentation. I have created the model but I am having difficulties in loading the dataset as I only have experience loading images with folder names as labels for image classification.

Will appreciate some help here to guide me on how to load training images and ground truth labels in PyTorch.

Thank you

2

I am assuming there is a folder which contains sub-folders image and mask. The filenames are same. This is a template you can modify it and use. All the best.

import torch.utils.data as data

class DataLoaderSegmentation(data.Dataset):
    def __init__(self, folder_path):
        super(DataLoaderSegmentation, self).__init__()
        self.img_files = glob.glob(os.path.join(folder_path,'image','*.png')
        self.mask_files = []
        for img_path in img_files:
             self.mask_files.append(os.path.join(folder_path,'mask',os.path.basename(img_path)) 

    def __getitem__(self, index):
            img_path = self.img_files[index]
            mask_path = self.mask_files[index]
            data = use opencv or pil read image using img_path
            label =use opencv or pil read label  using mask_path
            return torch.from_numpy(data).float(), torch.from_numpy(label).float()

    def __len__(self):
        return len(self.img_files)

I have already this in other thread. But I don’t know how to share that url. See if it helps