diff --git a/configs/_base_/models/deeplabv3_r50-d8.py b/configs/_base_/models/deeplabv3_r50-d8.py
index 725ae1afd0..36cc34947a 100644
--- a/configs/_base_/models/deeplabv3_r50-d8.py
+++ b/configs/_base_/models/deeplabv3_r50-d8.py
@@ -3,8 +3,8 @@
 norm_cfg = dict(type='SyncBN', requires_grad=True)
 data_preprocessor = dict(
     type='SegDataPreProcessor',
-    mean=[123.675, 116.28, 103.53],
-    std=[58.395, 57.12, 57.375],
+    mean=[0.4082, 0.3791, 0.2815],
+    std=[0.1351, 0.1022, 0.0931],
     bgr_to_rgb=True,
     pad_val=0,
     seg_pad_val=255)
diff --git a/configs/_base_/models/deeplabv3plus_r50-d8_deepglobe.py b/configs/_base_/models/deeplabv3plus_r50-d8_deepglobe.py
new file mode 100644
index 0000000000..1d63ef467f
--- /dev/null
+++ b/configs/_base_/models/deeplabv3plus_r50-d8_deepglobe.py
@@ -0,0 +1,54 @@
+# model settings
+norm_cfg = dict(type='SyncBN', requires_grad=True)
+data_preprocessor = dict(
+    type='SegDataPreProcessor',
+    mean=[0.4082, 0.3791, 0.2815],
+    std=[0.1351, 0.1022, 0.0931],
+    bgr_to_rgb=True,
+    pad_val=0,
+    seg_pad_val=255)
+model = dict(
+    type='EncoderDecoder',
+    data_preprocessor=data_preprocessor,
+    pretrained='open-mmlab://resnet50_v1c',
+    backbone=dict(
+        type='ResNetV1c',
+        depth=50,
+        num_stages=4,
+        out_indices=(0, 1, 2, 3),
+        dilations=(1, 1, 2, 4),
+        strides=(1, 2, 1, 1),
+        norm_cfg=norm_cfg,
+        norm_eval=False,
+        style='pytorch',
+        contract_dilation=True),
+    decode_head=dict(
+        type='DepthwiseSeparableASPPHead',
+        in_channels=2048,
+        in_index=3,
+        channels=512,
+        dilations=(1, 12, 24, 36),
+        c1_in_channels=256,
+        c1_channels=48,
+        dropout_ratio=0.1,
+        num_classes=7,
+        norm_cfg=norm_cfg,
+        align_corners=False,
+        loss_decode=dict(
+            type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
+    auxiliary_head=dict(
+        type='FCNHead',
+        in_channels=1024,
+        in_index=2,
+        channels=256,
+        num_convs=1,
+        concat_input=False,
+        dropout_ratio=0.1,
+        num_classes=7,
+        norm_cfg=norm_cfg,
+        align_corners=False,
+        loss_decode=dict(
+            type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
+    # model training and testing settings
+    train_cfg=dict(),
+    test_cfg=dict(mode='whole'))
diff --git a/configs/_base_/models/pspnet_unet_deepglobe_s5-d16.py b/configs/_base_/models/pspnet_unet_deepglobe_s5-d16.py
index f14a2ca089..8cefb930fe 100644
--- a/configs/_base_/models/pspnet_unet_deepglobe_s5-d16.py
+++ b/configs/_base_/models/pspnet_unet_deepglobe_s5-d16.py
@@ -2,8 +2,8 @@
 norm_cfg = dict(type='SyncBN', requires_grad=True)
 data_preprocessor = dict(
     type='SegDataPreProcessor',
-    mean=[123.675, 116.28, 103.53],
-    std=[58.395, 57.12, 57.375],
+    mean=[0.4082, 0.3791, 0.2815],
+    std=[0.1351, 0.1022, 0.0931],
     bgr_to_rgb=True,
     pad_val=0,
     seg_pad_val=255)
@@ -35,14 +35,14 @@
         channels=16,
         pool_scales=(1, 2, 3, 6),
         dropout_ratio=0.1,
-        num_classes=2,
+        num_classes=7,
         norm_cfg=norm_cfg,
         align_corners=False,
         loss_decode=dict(
             type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
     auxiliary_head=dict(
         type='FCNHead',
-        in_channels=1024,
+        in_channels=128,
         in_index=3,
         channels=256,
         num_convs=1,
diff --git a/configs/deeplabv3plus/deeplabv3plus_r50-d8_4xb2-40k_deepglobe-512x1024.py b/configs/deeplabv3plus/deeplabv3plus_r50-d8_4xb2-40k_deepglobe-512x1024.py
new file mode 100644
index 0000000000..bbd63a3be1
--- /dev/null
+++ b/configs/deeplabv3plus/deeplabv3plus_r50-d8_4xb2-40k_deepglobe-512x1024.py
@@ -0,0 +1,8 @@
+_base_ = [
+    '../_base_/models/deeplabv3plus_r50-d8_deepglobe.py',
+    '../_base_/datasets/deepglobe.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_40k.py'
+]
+crop_size = (256, 256)
+data_preprocessor = dict(size=crop_size)
+model = dict(data_preprocessor=data_preprocessor)