1.インストール
最適な環境は Ubuntu (Linux) ですが、Windows では解決がより困難な問題が発生します。
https://github.com/open-mmlab/mmsegmentation/blob/master/docs/en/get_started.md#installation
リンクを開いた後、最初の方法を使用してください。これは後で直接使用されます。実際、同時に mmsegmentation を pip install することもできます。以下は参考までに私が作成した仮想環境です。
# Name Version Build Channel
_libgcc_mutex 0.1 main defaults
_openmp_mutex 5.1 1_gnu defaults
addict 2.4.0 pypi_0 pypi
anykeystore 0.2 pypi_0 pypi
apex 0.1 pypi_0 pypi
blas 1.0 mkl defaults
ca-certificates 2022.4.26 h06a4308_0 defaults
certifi 2022.6.15 py38h06a4308_0 defaults
charset-normalizer 2.0.12 pypi_0 pypi
click 8.1.3 pypi_0 pypi
colorama 0.4.5 pypi_0 pypi
commonmark 0.9.1 pypi_0 pypi
cryptacular 1.6.2 pypi_0 pypi
cycler 0.11.0 pypi_0 pypi
defusedxml 0.7.1 pypi_0 pypi
fonttools 4.33.3 pypi_0 pypi
greenlet 1.1.2 pypi_0 pypi
hupper 1.10.3 pypi_0 pypi
idna 3.3 pypi_0 pypi
importlib-metadata 4.12.0 pypi_0 pypi
intel-openmp 2021.4.0 h06a4308_3561 defaults
kiwisolver 1.4.3 pypi_0 pypi
ld_impl_linux-64 2.38 h1181459_1 defaults
libffi 3.3 he6710b0_2 defaults
libgcc-ng 11.2.0 h1234567_1 defaults
libgfortran-ng 7.5.0 ha8ba4b0_17 defaults
libgfortran4 7.5.0 ha8ba4b0_17 defaults
libgomp 11.2.0 h1234567_1 defaults
libstdcxx-ng 11.2.0 h1234567_1 defaults
markdown 3.3.7 pypi_0 pypi
markupsafe 2.1.1 pypi_0 pypi
matplotlib 3.5.2 pypi_0 pypi
mkl 2021.4.0 h06a4308_640 defaults
mkl-service 2.4.0 py38h7f8727e_0 defaults
mkl_fft 1.3.1 py38hd3c417c_0 defaults
mkl_random 1.2.2 py38h51133e4_0 defaults
mmcls 0.23.1 pypi_0 pypi
mmcv-full 1.5.3 pypi_0 pypi
mmsegmentation 0.25.0 pypi_0 pypi
model-index 0.1.11 pypi_0 pypi
ncurses 6.3 h7f8727e_2 defaults
numpy 1.23.0 pypi_0 pypi
numpy-base 1.22.3 py38hf524024_0 defaults
oauthlib 3.2.0 pypi_0 pypi
opencv-python 4.6.0.66 pypi_0 pypi
openmim 0.2.0 pypi_0 pypi
openssl 1.1.1o h7f8727e_0 defaults
ordered-set 4.1.0 pypi_0 pypi
packaging 21.3 pypi_0 pypi
pandas 1.4.3 pypi_0 pypi
pastedeploy 2.1.1 pypi_0 pypi
pbkdf2 1.3 pypi_0 pypi
pillow 9.1.1 pypi_0 pypi
pip 21.2.4 py38h06a4308_0 defaults
plaster 1.0 pypi_0 pypi
plaster-pastedeploy 0.7 pypi_0 pypi
prettytable 3.3.0 pypi_0 pypi
protobuf 3.20.1 pypi_0 pypi
pygments 2.12.0 pypi_0 pypi
pyparsing 3.0.9 pypi_0 pypi
pyramid 2.0 pypi_0 pypi
pyramid-mailer 0.15.1 pypi_0 pypi
python 3.8.13 h12debd9_0 defaults
python-dateutil 2.8.2 pypi_0 pypi
python3-openid 3.2.0 pypi_0 pypi
pytz 2022.1 pypi_0 pypi
pyyaml 6.0 pypi_0 pypi
readline 8.1.2 h7f8727e_1 defaults
repoze-sendmail 4.4.1 pypi_0 pypi
requests 2.28.0 pypi_0 pypi
requests-oauthlib 1.3.1 pypi_0 pypi
rich 12.4.4 pypi_0 pypi
scipy 1.7.3 py38hc147768_0 defaults
setuptools 61.2.0 py38h06a4308_0 defaults
six 1.16.0 pyhd3eb1b0_1 defaults
sqlalchemy 1.4.39 pypi_0 pypi
sqlite 3.38.5 hc218d9a_0 defaults
tabulate 0.8.10 pypi_0 pypi
tensorboardx 2.5.1 pypi_0 pypi
timm 0.3.2 pypi_0 pypi
tk 8.6.12 h1ccaba5_0 defaults
torch 1.8.0+cu111 pypi_0 pypi
torchvision 0.9.0+cu111 pypi_0 pypi
transaction 3.0.1 pypi_0 pypi
translationstring 1.4 pypi_0 pypi
typing-extensions 4.2.0 pypi_0 pypi
urllib3 1.26.9 pypi_0 pypi
velruse 1.1.1 pypi_0 pypi
venusian 3.0.0 pypi_0 pypi
wcwidth 0.2.5 pypi_0 pypi
webob 1.8.7 pypi_0 pypi
wheel 0.37.1 pyhd3eb1b0_0 defaults
wtforms 3.0.1 pypi_0 pypi
wtforms-recaptcha 0.3.2 pypi_0 pypi
xz 5.2.5 h7f8727e_1 defaults
yapf 0.32.0 pypi_0 pypi
zipp 3.8.0 pypi_0 pypi
zlib 1.2.12 h7f8727e_2 defaults
zope-deprecation 4.4.0 pypi_0 pypi
zope-interface 5.4.0 pypi_0 pypi
zope-sqlalchemy 1.6 pypi_0 pypi2. トレーニング
以下は公式が提供しているサンプルですが、実際にこれに従えばツール全体のロジックをマスターすることができます。
https://github.com/open-mmlab/mmsegmentation/blob/master/demo/MMSegmentation_Tutorial.ipynb
データダウンロード、データも上のリンクにあります
データをダウンロードしたら、忘れずに標準形式に変換してください。
main.py ファイルとトレーニング スクリプトを作成します。これらも上記のリンクから派生したものです。コードのロジックは、よく見ると非常に明確です。最初にデータの準備、次に構成ファイルの変更、変更後のビルドが行われます。モデル、そして最後にトレーニングです。
import mmcv
import os.path as osp
from PIL import Image
import matplotlib.pyplot as plt
import matplotlib.patches as mpatches
import numpy as np
from mmseg.datasets.builder import DATASETS
from mmseg.datasets.custom import CustomDataset
from mmcv import Config
from mmseg.apis import set_random_seed
from mmseg.datasets import build_dataset
from mmseg.models import build_segmentor
from mmseg.apis import train_segmentor
import os
os.environ['KMP_DUPLICATE_LIB_OK'] = 'True'
data_root = './stanford_background/iccv09Data'
img_dir = 'images'
ann_dir = 'labels'
classes = ('sky', 'tree', 'road', 'grass', 'water', 'bldg', 'mntn', 'fg obj')
palette = [[128, 128, 128], [129, 127, 38], [120, 69, 125], [53, 125, 34],
[0, 11, 123], [118, 20, 12], [122, 81, 25], [241, 134, 51]]
@DATASETS.register_module()
class StanfordBackgroundDataset(CustomDataset):
CLASSES = classes
PALETTE = palette
def __init__(self, split, **kwargs):
super().__init__(img_suffix='.jpg', seg_map_suffix='.png',
split=split, **kwargs)
assert osp.exists(self.img_dir) and self.split is not None
# split train/val set randomly
split_dir = 'splits'
mmcv.mkdir_or_exist(osp.join(data_root, split_dir))
filename_list = [osp.splitext(filename)[0] for filename in mmcv.scandir(
osp.join(data_root, ann_dir), suffix='.png')]
with open(osp.join(data_root, split_dir, 'train.txt'), 'w') as f:
# select first 4/5 as train set
train_length = int(len(filename_list)*4/5)
f.writelines(line + '\n' for line in filename_list[:train_length])
with open(osp.join(data_root, split_dir, 'val.txt'), 'w') as f:
# select last 1/5 as train set
f.writelines(line + '\n' for line in filename_list[train_length:])
############################################################################################ 这里训练pspnet网络，通过在脚本里修改配置文件，里面还有很多参数，想改别的打开看看细节吧
cfg = Config.fromfile('./configs/pspnet/pspnet_r50-d8_512x1024_40k_cityscapes.py')
# Since we use only one GPU, BN is used instead of SyncBN
cfg.norm_cfg = dict(type='BN', requires_grad=True)
cfg.model.backbone.norm_cfg = cfg.norm_cfg
cfg.model.decode_head.norm_cfg = cfg.norm_cfg
cfg.model.auxiliary_head.norm_cfg = cfg.norm_cfg
# modify num classes of the model in decode/auxiliary head
cfg.model.decode_head.num_classes = 8
cfg.model.auxiliary_head.num_classes = 8
# Modify dataset type and path
cfg.dataset_type = 'StanfordBackgroundDataset'
cfg.data_root = data_root
cfg.data.samples_per_gpu = 8
cfg.data.workers_per_gpu = 1
cfg.img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
cfg.crop_size = (256, 256)
cfg.train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='LoadAnnotations'),
dict(type='Resize', img_scale=(320, 240), ratio_range=(0.5, 2.0)),
dict(type='RandomCrop', crop_size=cfg.crop_size, cat_max_ratio=0.75),
dict(type='RandomFlip', flip_ratio=0.5),
dict(type='PhotoMetricDistortion'),
dict(type='Normalize', **cfg.img_norm_cfg),
dict(type='Pad', size=cfg.crop_size, pad_val=0, seg_pad_val=255),
dict(type='DefaultFormatBundle'),
dict(type='Collect', keys=['img', 'gt_semantic_seg']),
]
cfg.test_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='MultiScaleFlipAug',
img_scale=(320, 240),
# img_ratios=[0.5, 0.75, 1.0, 1.25, 1.5, 1.75],
flip=False,
transforms=[
dict(type='Resize', keep_ratio=True),
dict(type='RandomFlip'),
dict(type='Normalize', **cfg.img_norm_cfg),
dict(type='ImageToTensor', keys=['img']),
dict(type='Collect', keys=['img']),
])
]
cfg.data.train.type = cfg.dataset_type
cfg.data.train.data_root = cfg.data_root
cfg.data.train.img_dir = img_dir
cfg.data.train.ann_dir = ann_dir
cfg.data.train.pipeline = cfg.train_pipeline
cfg.data.train.split = 'splits/train.txt'
cfg.data.val.type = cfg.dataset_type
cfg.data.val.data_root = cfg.data_root
cfg.data.val.img_dir = img_dir
cfg.data.val.ann_dir = ann_dir
cfg.data.val.pipeline = cfg.test_pipeline
cfg.data.val.split = 'splits/val.txt'
cfg.data.test.type = cfg.dataset_type
cfg.data.test.data_root = cfg.data_root
cfg.data.test.img_dir = img_dir
cfg.data.test.ann_dir = ann_dir
cfg.data.test.pipeline = cfg.test_pipeline
cfg.data.test.split = 'splits/val.txt'
# We can still use the pre-trained Mask RCNN model though we do not need to
# use the mask branch
#预模型位置，链接里有下载地址，自己去下载一下
cfg.load_from = 'checkpoints/pspnet_r50-d8_512x1024_40k_cityscapes_20200605_003338-2966598c.pth'
# Set up working dir to save files and logs.
#训练生成的文件位置
cfg.work_dir = './run/pspnet/'
cfg.runner.max_iters = 100000
cfg.log_config.interval = 20
cfg.evaluation.interval = 500
cfg.checkpoint_config.interval = 200
# Set seed to facitate reproducing the result
cfg.seed = 0
set_random_seed(0, deterministic=False)
cfg.gpu_ids = range(1)
# Let's have a look at the final config used for training
print(f'Config:\n{cfg.pretty_text}')
# Build the dataset
datasets = [build_dataset(cfg.data.train)]
# Build the detector
model = build_segmentor(cfg.model)
# Add an attribute for visualization convenience
model.CLASSES = datasets[0].CLASSES
# Create work_dir
mmcv.mkdir_or_exist(osp.abspath(cfg.work_dir))
train_segmentor(model, datasets, cfg, distributed=False, validate=True,
meta=dict())
上記の操作後、main.py を直接実行してトレーニングを開始できるので、非常に便利です。
3. 予測
以下は、私が使用する予測コード、カテゴリとカラーマップが存在する必要がある、および構成ファイルです。
import mmcv
import os.path as osp
from PIL import Image
import matplotlib.pyplot as plt
import matplotlib.patches as mpatches
import numpy as np
from mmseg.datasets.builder import DATASETS
from mmseg.datasets.custom import CustomDataset
from mmcv import Config
from mmseg.apis import set_random_seed
from mmseg.datasets import build_dataset
from mmseg.models import build_segmentor
from mmseg.apis import train_segmentor, inference_segmentor, init_segmentor, show_result_pyplot
import os
import cv2
os.environ['KMP_DUPLICATE_LIB_OK'] = 'True'
data_root = './stanford_background/iccv09Data'
img_dir = 'images'
ann_dir = 'labels'
classes = ('sky', 'tree', 'road', 'grass', 'water', 'bldg', 'mntn', 'fg obj')
palette = [[128, 128, 128], [129, 127, 38], [120, 69, 125], [53, 125, 34],
[0, 11, 123], [118, 20, 12], [122, 81, 25], [241, 134, 51]]
@DATASETS.register_module()
class StanfordBackgroundDataset(CustomDataset):
CLASSES = classes
PALETTE = palette
def __init__(self, split, **kwargs):
super().__init__(img_suffix='.jpg', seg_map_suffix='.png',
split=split, **kwargs)
assert osp.exists(self.img_dir) and self.split is not None
############################################################################################
cfg = Config.fromfile('./configs/pspnet/pspnet_r50-d8_512x1024_40k_cityscapes.py')
# Since we use only one GPU, BN is used instead of SyncBN
cfg.norm_cfg = dict(type='BN', requires_grad=True)
cfg.model.backbone.norm_cfg = cfg.norm_cfg
cfg.model.decode_head.norm_cfg = cfg.norm_cfg
cfg.model.auxiliary_head.norm_cfg = cfg.norm_cfg
# modify num classes of the model in decode/auxiliary head
cfg.model.decode_head.num_classes = 8
cfg.model.auxiliary_head.num_classes = 8
# Modify dataset type and path
cfg.dataset_type = 'StanfordBackgroundDataset'
cfg.data_root = data_root
cfg.data.samples_per_gpu = 2
cfg.data.workers_per_gpu = 0
cfg.img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
cfg.crop_size = (256, 256)
cfg.train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='LoadAnnotations'),
dict(type='Resize', img_scale=(320, 240), ratio_range=(0.5, 2.0)),
dict(type='RandomCrop', crop_size=cfg.crop_size, cat_max_ratio=0.75),
dict(type='RandomFlip', flip_ratio=0.5),
dict(type='PhotoMetricDistortion'),
dict(type='Normalize', **cfg.img_norm_cfg),
dict(type='Pad', size=cfg.crop_size, pad_val=0, seg_pad_val=255),
dict(type='DefaultFormatBundle'),
dict(type='Collect', keys=['img', 'gt_semantic_seg']),
]
cfg.test_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='MultiScaleFlipAug',
img_scale=(320, 240),
# img_ratios=[0.5, 0.75, 1.0, 1.25, 1.5, 1.75],
flip=False,
transforms=[
dict(type='Resize', keep_ratio=True),
dict(type='RandomFlip'),
dict(type='Normalize', **cfg.img_norm_cfg),
dict(type='ImageToTensor', keys=['img']),
dict(type='Collect', keys=['img']),
])
]
cfg.data.train.type = cfg.dataset_type
cfg.data.train.data_root = cfg.data_root
cfg.data.train.img_dir = img_dir
cfg.data.train.ann_dir = ann_dir
cfg.data.train.pipeline = cfg.train_pipeline
cfg.data.train.split = 'splits/train.txt'
cfg.data.val.type = cfg.dataset_type
cfg.data.val.data_root = cfg.data_root
cfg.data.val.img_dir = img_dir
cfg.data.val.ann_dir = ann_dir
cfg.data.val.pipeline = cfg.test_pipeline
cfg.data.val.split = 'splits/val.txt'
cfg.data.test.type = cfg.dataset_type
cfg.data.test.data_root = cfg.data_root
cfg.data.test.img_dir = img_dir
cfg.data.test.ann_dir = ann_dir
cfg.data.test.pipeline = cfg.test_pipeline
cfg.data.test.split = 'splits/val.txt'
# We can still use the pre-trained Mask RCNN model though we do not need to
# use the mask branch
cfg.load_from = 'checkpoints/pspnet_r50-d8_512x1024_40k_cityscapes_20200605_003338-2966598c.pth'
# Set up working dir to save files and logs.
cfg.work_dir = './run'
cfg.runner.max_iters = 200
cfg.log_config.interval = 10
cfg.evaluation.interval = 200
cfg.checkpoint_config.interval = 200
# Set seed to facitate reproducing the result
cfg.seed = 0
set_random_seed(0, deterministic=False)
cfg.gpu_ids = range(1)
# Let's have a look at the final config used for training
print(f'Config:\n{cfg.pretty_text}')
config_file = cfg
checkpoints_file = './run/pspnet/latest.pth'
model = init_segmentor(config_file, checkpoints_file, device='cuda:0')
img = mmcv.imread('./stanford_background/iccv09Data/images/6000124.jpg')
result = inference_segmentor(model, img)
print(result)
#plt.figure(figsize=(8, 6))
# show_result_pyplot(model, img, result, palette)
#model.show_result(img, result, show=True)
cv2.imwrite('./re.jpg', result[0])
