CodeCamp #139 [Feature] Support REFUGE dataset.

configs/_base_/datasets/refuge.py

@@ -4,54 +4,38 @@
 train_img_scale = (2056, 2124)
 val_img_scale = (1634, 1634)
 test_img_scale = (1634, 1634)
-crop_size = (256, 256)
+crop_size = (512, 512)
+
 train_pipeline = [
     dict(type='LoadImageFromFile'),
-    dict(type='LoadAnnotations', reduce_zero_label=True),
+    dict(type='LoadAnnotations', reduce_zero_label=False),
     dict(
         type='RandomResize',
         scale=train_img_scale,
         ratio_range=(0.5, 2.0),
         keep_ratio=True),
     dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75),
-    dict(type='RandomFliplr', prob=0.2),
-    dict(type='RandomFlipud', prob=0.2),
-    dict(
-        type='RandomApply',
-        transforms=[dict(type='Rot90', degree_range=(1, 3))],
-        prob=0.3),
-    dict(
-        type='RandomChoice',
-        transforms=[
-            dict(type='ColorJitter', brightness=0.2),
-            dict(type='ColorJitter', contrast=0.2),
-            dict(type='ColorJitter', saturation=0.2),
-            dict(
-                type='ColorJitter',
-                brightness=0.1,
-                contrast=0.1,
-                saturation=0.1,
-                hue=0)
-        ]),
+    dict(type='RandomFlip', prob=0.5),
+    dict(type='PhotoMetricDistortion'),
     dict(type='PackSegInputs')
 ]
-
 val_pipeline = [
     dict(type='LoadImageFromFile'),
     dict(type='Resize', scale=val_img_scale, keep_ratio=True),
     # add loading annotation after ``Resize`` because ground truth
     # does not need to do resize data transform
-    dict(type='LoadAnnotations', reduce_zero_label=True),
+    dict(type='LoadAnnotations', reduce_zero_label=False),
     dict(type='PackSegInputs')
 ]
 test_pipeline = [
     dict(type='LoadImageFromFile'),
     dict(type='Resize', scale=test_img_scale, keep_ratio=True),
     # add loading annotation after ``Resize`` because ground truth
     # does not need to do resize data transform
-    dict(type='LoadAnnotations', reduce_zero_label=True),
+    dict(type='LoadAnnotations', reduce_zero_label=False),
     dict(type='PackSegInputs')
 ]
+
 train_dataloader = dict(
     batch_size=4,
     num_workers=4,

docs/en/user_guides/2_dataset_prepare.md

@@ -138,6 +138,15 @@ mmsegmentation
 │   │   ├── ann_dir
 │   │   │   ├── train
 │   │   │   ├── val
+│   ├── REFUGE
+│   │   ├── images
+│   │   │   ├── training
+│   │   │   ├── validation
+│   │   │   ├── test
+│   │   ├── annotations
+│   │   │   ├── training
+│   │   │   ├── validation
+│   │   │   ├── test
 ```
 
 ### Cityscapes
@@ -264,18 +273,6 @@ python tools/dataset_converters/hrf.py /path/to/healthy.zip /path/to/healthy_man
 
 The script will make directory structure automatically.
 
-### REFUGE
-
-First, registrate in [REFUGE2020Challenge](https://refuge.grand-challenge.org)and download data from [this link](https://refuge.grand-challenge.org/REFUGE2Download)
-
-To convert REFUGE dataset to MMSegmentation format, you should run the following command:
-
-```shell
-python tools/convert_datasets/refuge.py --raw_data_root=.. --tmp_dir=..
-```
-
-The script will make directory structure automatically.
-
 ### STARE
 
 First, download [stare-images.tar](http://cecas.clemson.edu/~ahoover/stare/probing/stare-images.tar), [labels-ah.tar](http://cecas.clemson.edu/~ahoover/stare/probing/labels-ah.tar) and [labels-vk.tar](http://cecas.clemson.edu/~ahoover/stare/probing/labels-vk.tar).
@@ -507,3 +504,31 @@ python tools/dataset_converters/synapse.py --dataset-path /path/to/synapse
 
 Noted that MMSegmentation default evaluation metric (such as mean dice value) is calculated on 2D slice image,
 which is not comparable to results of 3D scan in some paper such as [TransUNet](https://arxiv.org/abs/2102.04306).
+
+### REFUGE
+
+First, registrate in [REFUGE2020Challenge](https://refuge.grand-challenge.org)and download data from [this link](https://refuge.grand-challenge.org/REFUGE2Download)
+
+Then, unzip `REFUGE2.zip` and the contents of original datasets include:
+
+```none
+├── REFUGE2
+│   ├── REFUGE2
+│   │   ├── Annotation-Training400.zip
+│   │   ├── REFUGE-Test400.zip
+│   │   ├── REFUGE-Test-GT.zip
+│   │   ├── REFUGE-Training400.zip
+│   │   ├── REFUGE-Validation400.zip
+│   │   ├── REFUGE-Validation400-GT.zip
+│   ├── __MACOSX
+```
+
+To convert REFUGE dataset to MMSegmentation format, you should run the following command:
+
+```shell
+python tools/convert_datasets/refuge.py --raw_data_root=/path/to/refuge/REFUGE2/REFUGE2
+```
+
+The script will make directory structure automatically.
+
+In our default setting, it will generate 400 images for training, 400 images for validation and 400 images for testing.

docs/zh_cn/user_guides/2_dataset_prepare.md

@@ -119,6 +119,15 @@ mmsegmentation
 │   │   ├── ann_dir
 │   │   │   ├── train
 │   │   │   ├── val
+│   ├── REFUGE
+│   │   ├── images
+│   │   │   ├── training
+│   │   │   ├── validation
+│   │   │   ├── test
+│   │   ├── annotations
+│   │   │   ├── training
+│   │   │   ├── validation
+│   │   │   ├── test
 ```
 
 ### Cityscapes
@@ -206,19 +215,6 @@ python tools/convert_datasets/hrf.py /path/to/healthy.zip /path/to/healthy_manua
 
 这个脚本将自动生成正确的文件夹结构。
 
-### REFUGE
-
-首先，去[REFUGE2020挑战赛官网](https://refuge.grand-challenge.org)注册后到指定[链接](https://refuge.grand-challenge.org/REFUGE2Download)
-下载数据,
-数据包含2018年的训练集、验证集和测试集。
-为了将 REFUGE 数据集转换成 MMSegmentation 格式，您需要运行如下命令：
-
-```shell
-python tools/convert_datasets/refuge.py --raw_data_root=.. --tmp_dir=..
-```
-
-这个脚本将自动生成正确的文件夹结构。
-
 ### STARE
 
 首先，下载 [stare-images.tar](http://cecas.clemson.edu/~ahoover/stare/probing/stare-images.tar), [labels-ah.tar](http://cecas.clemson.edu/~ahoover/stare/probing/labels-ah.tar) 和 [labels-vk.tar](http://cecas.clemson.edu/~ahoover/stare/probing/labels-vk.tar) 。
@@ -330,3 +326,33 @@ python tools/convert_datasets/isaid.py /path/to/iSAID
 ```
 
 使用我们默认的配置 (`patch_width`=896, `patch_height`=896,　`overlap_area`=384)， 将生成 33978 张图片的训练集和 11644 张图片的验证集。
+
+### REFUGE
+
+首先，去[REFUGE2020挑战赛官网](https://refuge.grand-challenge.org)注册后到指定[链接](https://refuge.grand-challenge.org/REFUGE2Download)
+下载数据,
+数据包含2018年的训练集、验证集和测试集。
+
+随后解压 `REFUGE2.zip` 然后得到如下所示的原始数据目录.
+
+```none
+├── REFUGE2
+│   ├── REFUGE2
+│   │   ├── Annotation-Training400.zip
+│   │   ├── REFUGE-Test400.zip
+│   │   ├── REFUGE-Test-GT.zip
+│   │   ├── REFUGE-Training400.zip
+│   │   ├── REFUGE-Validation400.zip
+│   │   ├── REFUGE-Validation400-GT.zip
+│   ├── __MACOSX
+```
+
+为了将 REFUGE 数据集转换成 MMSegmentation 格式，您需要运行如下命令：
+
+```shell
+python tools/convert_datasets/refuge.py --raw_data_root=/path/to/refuge/REFUGE2/REFUGE2
+```
+
+这个脚本将自动生成正确的文件夹结构。
+
+使用我们默认的配置， 将生成 400 张图片的训练集， 400 张图片的验证集和 400 张图片的测试集.

mmseg/datasets/__init__.py

@@ -17,6 +17,7 @@
 from .night_driving import NightDrivingDataset
 from .pascal_context import PascalContextDataset, PascalContextDataset59
 from .potsdam import PotsdamDataset
+from .refuge import REFUGEDataset
 from .stare import STAREDataset
 from .synapse import SynapseDataset
 # yapf: disable
@@ -48,5 +49,5 @@
     'DecathlonDataset', 'LIPDataset', 'ResizeShortestEdge',
     'BioMedicalGaussianNoise', 'BioMedicalGaussianBlur',
     'BioMedicalRandomGamma', 'BioMedical3DPad', 'RandomRotFlip',
-    'SynapseDataset'
+    'SynapseDataset', 'REFUGEDataset'
 ]

mmseg/datasets/refuge.py

@@ -14,13 +14,13 @@ class REFUGEDataset(BaseSegDataset):
     '.png'.
     """
     METAINFO = dict(
-        classes=(' Optic Cup', 'Optic Disc'),
-        palette=[[6, 230, 230], [56, 59, 120]])
+        classes=('background', ' Optic Cup', 'Optic Disc'),
+        palette=[[120, 120, 120], [6, 230, 230], [56, 59, 120]])
 
     def __init__(self, **kwargs) -> None:
         super().__init__(
             img_suffix='.png',
             seg_map_suffix='.png',
-            reduce_zero_label=True,
+            reduce_zero_label=False,
             **kwargs)
         assert self.file_client.exists(self.data_prefix['img_path'])

tests/test_datasets/test_dataset.py

@@ -9,7 +9,8 @@
 from mmseg.datasets import (ADE20KDataset, BaseSegDataset, CityscapesDataset,
                             COCOStuffDataset, DecathlonDataset, ISPRSDataset,
                             LIPDataset, LoveDADataset, PascalVOCDataset,
-                            PotsdamDataset, SynapseDataset, iSAIDDataset)
+                            PotsdamDataset, REFUGEDataset, SynapseDataset,
+                            iSAIDDataset)
 from mmseg.registry import DATASETS
 from mmseg.utils import get_classes, get_palette
 
@@ -233,6 +234,19 @@ def test_synapse():
     assert len(test_dataset) == 2
 
 
+def test_refuge():
+    test_dataset = REFUGEDataset(
+        pipeline=[],
+        data_prefix=dict(
+            img_path=osp.join(
+                osp.dirname(__file__),
+                '../data/pseudo_refuge_dataset/img_dir'),
+            seg_map_path=osp.join(
+                osp.dirname(__file__),
+                '../data/pseudo_refuge_dataset/ann_dir')))
+    assert len(test_dataset) == 1
+
+
 def test_isaid():
     test_dataset = iSAIDDataset(
         pipeline=[],