ade.py

# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import os

from .dataset import Dataset
from utils.download import download_file_and_uncompress

DATA_HOME = os.path.expanduser('~/.cache/paddle/dataset')
URL = "https://paddleseg.bj.bcebos.com/dataset/ADEChallengeData2016.zip"


class ADE20K(Dataset):
    """ADE20K dataset `http://sceneparsing.csail.mit.edu/`.
    Args:
        data_dir: The dataset directory.
        mode: Which part of dataset to use.. it is one of ('train', 'val'). Default: 'train'.
        transforms: Transforms for image.
        download: Whether to download dataset if data_dir is None.
    """

    def __init__(self,
                 data_dir=None,
                 mode='train',
                 transforms=None,
                 download=True):
        self.data_dir = data_dir
        self.transforms = transforms
        self.mode = mode
        self.file_list = list()
        self.num_classes = 21

        if mode.lower() not in ['train', 'val']:
            raise Exception(
                "mode should be one of ('train', 'val') in PascalVOC dataset, but got {}."
                .format(mode))

        if self.transforms is None:
            raise Exception("transforms is necessary, but it is None.")

        if self.data_dir is None:
            if not download:
                raise Exception("data_dir not set and auto download disabled.")
            self.data_dir = download_file_and_uncompress(
                url=URL,
                savepath=DATA_HOME,
                extrapath=DATA_HOME,
                extraname='ADEChallengeData2016')

        if mode == 'train':
            img_dir = os.path.join(self.data_dir, 'images/training')
            grt_dir = os.path.join(self.data_dir, 'annotations/training')
        elif mode == 'val':
            img_dir = os.path.join(self.data_dir, 'images/validation')
            grt_dir = os.path.join(self.data_dir, 'annotations/validation')
        img_files = os.listdir(img_dir)
        grt_files = [i.replace('.jpg', '.png') for i in img_files]
        for i in range(len(img_files)):
            img_path = os.path.join(img_dir, img_files[i])
            grt_path = os.path.join(grt_dir, grt_files[i])
            self.file_list.append([img_path, grt_path])