open-mmlab · xvjiarui · Sep 30, 2021 · Sep 8, 2021 · Sep 8, 2021 · Sep 8, 2021
diff --git a/README.md b/README.md
@@ -77,6 +77,7 @@ Supported methods:
 - [x] [PSANet (ECCV'2018)](configs/psanet)
 - [x] [DeepLabV3+ (CVPR'2018)](configs/deeplabv3plus)
 - [x] [UPerNet (ECCV'2018)](configs/upernet)
+- [x] [ICNet (ECCV'2018)](configs/icnet)
 - [x] [NonLocal Net (CVPR'2018)](configs/nonlocal_net)
 - [x] [EncNet (CVPR'2018)](configs/encnet)
 - [x] [Semantic FPN (CVPR'2019)](configs/sem_fpn)

diff --git a/README_zh-CN.md b/README_zh-CN.md
@@ -76,6 +76,7 @@ MMSegmentation 是一个基于 PyTorch 的语义分割开源工具箱。它是 O
 - [x] [PSANet (ECCV'2018)](configs/psanet)
 - [x] [DeepLabV3+ (CVPR'2018)](configs/deeplabv3plus)
 - [x] [UPerNet (ECCV'2018)](configs/upernet)
+- [x] [ICNet (ECCV'2018)](configs/icnet)
 - [x] [NonLocal Net (CVPR'2018)](configs/nonlocal_net)
 - [x] [EncNet (CVPR'2018)](configs/encnet)
 - [x] [Semantic FPN (CVPR'2019)](configs/sem_fpn)

diff --git a/configs/_base_/datasets/cityscapes_832x832.py b/configs/_base_/datasets/cityscapes_832x832.py
@@ -0,0 +1,35 @@
+_base_ = './cityscapes.py'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+crop_size = (832, 832)
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations'),
+    dict(type='Resize', img_scale=(2048, 1024), ratio_range=(0.5, 2.0)),
+    dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75),
+    dict(type='RandomFlip', prob=0.5),
+    dict(type='PhotoMetricDistortion'),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='Pad', size=crop_size, pad_val=0, seg_pad_val=255),
+    dict(type='DefaultFormatBundle'),
+    dict(type='Collect', keys=['img', 'gt_semantic_seg']),
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='MultiScaleFlipAug',
+        img_scale=(2048, 1024),
+        # 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', **img_norm_cfg),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(type='Collect', keys=['img']),
+        ])
+]
+data = dict(
+    train=dict(pipeline=train_pipeline),
+    val=dict(pipeline=test_pipeline),
+    test=dict(pipeline=test_pipeline))
diff --git a/configs/_base_/models/icnet_r50-d8.py b/configs/_base_/models/icnet_r50-d8.py
@@ -0,0 +1,41 @@
+# model settings
+norm_cfg = dict(type='SyncBN', requires_grad=True)
+model = dict(
+    type='EncoderDecoder',
+    backbone=dict(
+        type='ICNet',
+        in_channels=3,
+        layer_channels=(512, 2048),
+        light_branch_middle_channels=32,
+        psp_out_channels=512,
+        out_channels=(64, 256, 256),
+        backbone_cfg=dict(
+            type='ResNetV1c',
+            in_channels=3,
+            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),
+        norm_cfg=norm_cfg,
+        align_corners=False,
+    ),
+    decode_head=dict(
+        type='ICHead',
+        in_channels=(64, 256, 256),
+        in_index=(0, 1, 2),
+        input_transform='multiple_select',
+        channels=128,
+        dropout_ratio=0,
+        norm_cfg=norm_cfg,
+        align_corners=False,
+        num_classes=19,
+        loss_decode=dict(
+            type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
+    # model training and testing settings
+    train_cfg=dict(),
+    test_cfg=dict(mode='whole'))
diff --git a/configs/icnet/README.md b/configs/icnet/README.md
@@ -0,0 +1,36 @@
+# Icnet for real-time semantic segmentation on high-resolution images
+
+## Introduction
+
+<!-- [ALGORITHM] -->
+
+```latex
+@inproceedings{zhao2018icnet,
+  title={Icnet for real-time semantic segmentation on high-resolution images},
+  author={Zhao, Hengshuang and Qi, Xiaojuan and Shen, Xiaoyong and Shi, Jianping and Jia, Jiaya},
+  booktitle={Proceedings of the European conference on computer vision (ECCV)},
+  pages={405--420},
+  year={2018}
+}
+```
+
+## Results and models
+
+### Cityscapes
+
+| Method | Backbone   | Crop Size | Lr schd | Mem (GB) | Inf time (fps) |  mIoU | mIoU(ms+flip) | config                                                                                                                     | download                                                                                                                                                                                                                                                                                                                                       |
+| ------ | ---------- | --------- | ------: | -------- | -------------- | ----: | ------------: | -------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| ICNet    | R-18-D8    | 832x832  |   80000 | 1.87      | 32.11          | 66.92 |         69.43 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r18-d8_832x832_80k_cityscapes.py)       | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_512x1024_40k_cityscapes/fcn_r50-d8_512x1024_40k_cityscapes_20200604_192608-efe53f0d.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_512x1024_40k_cityscapes/fcn_r50-d8_512x1024_40k_cityscapes_20200604_192608.log.json)         |
+| ICNet    | R-18-D8   | 832x832  |   160000 | -     | -           | 71.77 |         74.71 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r18-d8_832x832_160k_cityscapes.py)      | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_512x1024_40k_cityscapes/fcn_r101-d8_512x1024_40k_cityscapes_20200604_181852-a883d3a1.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_512x1024_40k_cityscapes/fcn_r101-d8_512x1024_40k_cityscapes_20200604_181852.log.json)     |
+| ICNet (in1k-pre)    | R-18-D8    | 832x832   |   80000 | -      | -           | 73.16 |         75.31 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r18-d8_in1k-pre_832x832_80k_cityscapes.py)        | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_769x769_40k_cityscapes/fcn_r50-d8_769x769_40k_cityscapes_20200606_113104-977b5d02.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_769x769_40k_cityscapes/fcn_r50-d8_769x769_40k_cityscapes_20200606_113104.log.json)             |
+| ICNet (in1k-pre)    | R-18-D8   | 832x832   |   160000 | -     | -           | 71.28 |         73.58 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r18-d8_in1k-pre_832x832_160k_cityscapes.py)       | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_769x769_40k_cityscapes/fcn_r101-d8_769x769_40k_cityscapes_20200606_113208-7d4ab69c.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_769x769_40k_cityscapes/fcn_r101-d8_769x769_40k_cityscapes_20200606_113208.log.json)         |
+| ICNet    | R-50-D8    | 832x832  |   80000 | 2.41      | 19.46          | 69.66 |         71.69 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r50-d8_832x832_80k_cityscapes.py)       | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_512x1024_80k_cityscapes/fcn_r18-d8_512x1024_80k_cityscapes_20201225_021327-6c50f8b4.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_512x1024_80k_cityscapes/fcn_r18-d8_512x1024_80k_cityscapes-20201225_021327.log.json)         |
+| ICNet    | R-50-D8    | 832x832  |   160000 | -       | -               | 72.38 |         74.37 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r50-d8_832x832_160k_cityscapes.py)       | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_512x1024_80k_cityscapes/fcn_r50-d8_512x1024_80k_cityscapes_20200606_113019-03aa804d.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_512x1024_80k_cityscapes/fcn_r50-d8_512x1024_80k_cityscapes_20200606_113019.log.json)         |
+| ICNet (in1k-pre)    | R-50-D8   | 832x832  |   80000 | -        | -              | 74.1 |         76.52 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r50-d8_in1k-pre_832x832_80k_cityscapes.py)      | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_512x1024_80k_cityscapes/fcn_r101-d8_512x1024_80k_cityscapes_20200606_113038-3fb937eb.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_512x1024_80k_cityscapes/fcn_r101-d8_512x1024_80k_cityscapes_20200606_113038.log.json)     |
+| ICNet (in1k-pre)    | R-50-D8    | 832x832   |   160000 | -      | -           | 70.08 |         77.49 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r50-d8_in1k-pre_832x832_160k_cityscapes.py)        | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_769x769_80k_cityscapes/fcn_r18-d8_769x769_80k_cityscapes_20201225_021451-9739d1b8.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_769x769_80k_cityscapes/fcn_r18-d8_769x769_80k_cityscapes-20201225_021451.log.json)             |
+| ICNet    | R-101-D8    | 832x832  |   80000 | 3.26      | 15.11          | 70.34 |         72.39 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r101-d8_832x832_80k_cityscapes.py)       | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_512x1024_80k_cityscapes/fcn_r18-d8_512x1024_80k_cityscapes_20201225_021327-6c50f8b4.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_512x1024_80k_cityscapes/fcn_r18-d8_512x1024_80k_cityscapes-20201225_021327.log.json)         |
+| ICNet    | R-101-D8    | 832x832  |   160000 | -        | -               | 73.19 |         75.72 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r101-d8_832x832_160k_cityscapes.py)       | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_512x1024_80k_cityscapes/fcn_r50-d8_512x1024_80k_cityscapes_20200606_113019-03aa804d.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_512x1024_80k_cityscapes/fcn_r50-d8_512x1024_80k_cityscapes_20200606_113019.log.json)         |
+| ICNet (in1k-pre)    | R-101-D8   | 832x832  |   80000 | -       | -              | 75.19 |         77.82 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r101-d8_in1k-pre_832x832_80k_cityscapes.py)      | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_512x1024_80k_cityscapes/fcn_r101-d8_512x1024_80k_cityscapes_20200606_113038-3fb937eb.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_512x1024_80k_cityscapes/fcn_r101-d8_512x1024_80k_cityscapes_20200606_113038.log.json)     |
+| ICNet (in1k-pre)    | R-101-D8    | 832x832   |   160000 | -      | -           | 75.16 |         77.32 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/icnet/icnet_r101-d8_in1k-pre_832x832_160k_cityscapes.py)        | [model](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_769x769_80k_cityscapes/fcn_r18-d8_769x769_80k_cityscapes_20201225_021451-9739d1b8.pth) &#124; [log](https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_769x769_80k_cityscapes/fcn_r18-d8_769x769_80k_cityscapes-20201225_021451.log.json)             |
+
+Note: `in1k-pre` means pretrained model is used.
diff --git a/configs/icnet/icnet.yml b/configs/icnet/icnet.yml
@@ -0,0 +1,198 @@
+Collections:
+- Metadata:
+    Training Data:
+    - Cityscapes
+  Name: icnet
+Models:
+- Config: configs/icnet/icnet_r18-d8_832x832_80k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-18-D8
+    crop size: (832,832)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (832,832)
+      value: 31.14
+    lr schd: 80000
+    memory (GB): 1.87
+  Name: icnet_r18-d8_832x832_80k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 66.92
+      mIoU(ms+flip): 69.43
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_512x1024_40k_cityscapes/fcn_r50-d8_512x1024_40k_cityscapes_20200604_192608-efe53f0d.pth
+- Config: configs/icnet/icnet_r18-d8_832x832_160k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-18-D8
+    crop size: (832,832)
+    lr schd: 160000
+  Name: icnet_r18-d8_832x832_160k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 71.77
+      mIoU(ms+flip): 74.71
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_512x1024_40k_cityscapes/fcn_r101-d8_512x1024_40k_cityscapes_20200604_181852-a883d3a1.pth
+- Config: configs/icnet/icnet_r18-d8_in1k-pre_832x832_80k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-18-D8
+    crop size: (832,832)
+    lr schd: 80000
+  Name: icnet_r18-d8_in1k-pre_832x832_80k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 73.16
+      mIoU(ms+flip): 75.31
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_769x769_40k_cityscapes/fcn_r50-d8_769x769_40k_cityscapes_20200606_113104-977b5d02.pth
+- Config: configs/icnet/icnet_r18-d8_in1k-pre_832x832_160k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-18-D8
+    crop size: (832,832)
+    lr schd: 160000
+  Name: icnet_r18-d8_in1k-pre_832x832_160k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 71.28
+      mIoU(ms+flip): 73.58
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_769x769_40k_cityscapes/fcn_r101-d8_769x769_40k_cityscapes_20200606_113208-7d4ab69c.pth
+- Config: configs/icnet/icnet_r50-d8_832x832_80k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-50-D8
+    crop size: (832,832)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (832,832)
+      value: 51.39
+    lr schd: 80000
+    memory (GB): 2.41
+  Name: icnet_r50-d8_832x832_80k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 69.66
+      mIoU(ms+flip): 71.69
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_512x1024_80k_cityscapes/fcn_r18-d8_512x1024_80k_cityscapes_20201225_021327-6c50f8b4.pth
+- Config: configs/icnet/icnet_r50-d8_832x832_160k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-50-D8
+    crop size: (832,832)
+    lr schd: 160000
+  Name: icnet_r50-d8_832x832_160k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 72.38
+      mIoU(ms+flip): 74.37
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_512x1024_80k_cityscapes/fcn_r50-d8_512x1024_80k_cityscapes_20200606_113019-03aa804d.pth
+- Config: configs/icnet/icnet_r50-d8_in1k-pre_832x832_80k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-50-D8
+    crop size: (832,832)
+    lr schd: 80000
+  Name: icnet_r50-d8_in1k-pre_832x832_80k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 74.1
+      mIoU(ms+flip): 76.52
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_512x1024_80k_cityscapes/fcn_r101-d8_512x1024_80k_cityscapes_20200606_113038-3fb937eb.pth
+- Config: configs/icnet/icnet_r50-d8_in1k-pre_832x832_160k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-50-D8
+    crop size: (832,832)
+    lr schd: 160000
+  Name: icnet_r50-d8_in1k-pre_832x832_160k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 70.08
+      mIoU(ms+flip): 77.49
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_769x769_80k_cityscapes/fcn_r18-d8_769x769_80k_cityscapes_20201225_021451-9739d1b8.pth
+- Config: configs/icnet/icnet_r101-d8_832x832_80k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-101-D8
+    crop size: (832,832)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (832,832)
+      value: 66.18
+    lr schd: 80000
+    memory (GB): 3.26
+  Name: icnet_r101-d8_832x832_80k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 70.34
+      mIoU(ms+flip): 72.39
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_512x1024_80k_cityscapes/fcn_r18-d8_512x1024_80k_cityscapes_20201225_021327-6c50f8b4.pth
+- Config: configs/icnet/icnet_r101-d8_832x832_160k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-101-D8
+    crop size: (832,832)
+    lr schd: 160000
+  Name: icnet_r101-d8_832x832_160k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 73.19
+      mIoU(ms+flip): 75.72
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r50-d8_512x1024_80k_cityscapes/fcn_r50-d8_512x1024_80k_cityscapes_20200606_113019-03aa804d.pth
+- Config: configs/icnet/icnet_r101-d8_in1k-pre_832x832_80k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-101-D8
+    crop size: (832,832)
+    lr schd: 80000
+  Name: icnet_r101-d8_in1k-pre_832x832_80k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 75.19
+      mIoU(ms+flip): 77.82
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r101-d8_512x1024_80k_cityscapes/fcn_r101-d8_512x1024_80k_cityscapes_20200606_113038-3fb937eb.pth
+- Config: configs/icnet/icnet_r101-d8_in1k-pre_832x832_160k_cityscapes.py
+  In Collection: icnet
+  Metadata:
+    backbone: R-101-D8
+    crop size: (832,832)
+    lr schd: 160000
+  Name: icnet_r101-d8_in1k-pre_832x832_160k_cityscapes
+  Results:
+    Dataset: Cityscapes
+    Metrics:
+      mIoU: 75.16
+      mIoU(ms+flip): 77.32
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/fcn/fcn_r18-d8_769x769_80k_cityscapes/fcn_r18-d8_769x769_80k_cityscapes_20201225_021451-9739d1b8.pth
diff --git a/configs/icnet/icnet_r101-d8_832x832_160k_cityscapes.py b/configs/icnet/icnet_r101-d8_832x832_160k_cityscapes.py
@@ -0,0 +1,7 @@
+_base_ = [
+    '../_base_/models/icnet_r50-d8.py',
+    '../_base_/datasets/cityscapes_832x832.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_160k.py'
+]
+
+model = dict(backbone=dict(backbone_cfg=dict(depth=101)))
diff --git a/configs/icnet/icnet_r101-d8_832x832_80k_cityscapes.py b/configs/icnet/icnet_r101-d8_832x832_80k_cityscapes.py
@@ -0,0 +1,7 @@
+_base_ = [
+    '../_base_/models/icnet_r50-d8.py',
+    '../_base_/datasets/cityscapes_832x832.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_80k.py'
+]
+
+model = dict(backbone=dict(backbone_cfg=dict(depth=101)))
diff --git a/configs/icnet/icnet_r101-d8_in1k-pre_832x832_160k_cityscapes.py b/configs/icnet/icnet_r101-d8_in1k-pre_832x832_160k_cityscapes.py
@@ -0,0 +1,12 @@
+_base_ = [
+    '../_base_/models/icnet_r50-d8.py',
+    '../_base_/datasets/cityscapes_832x832.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_160k.py'
+]
+
+model = dict(
+    backbone=dict(
+        backbone_cfg=dict(
+            depth=101,
+            init_cfg=dict(
+                type='Pretrained', checkpoint='open-mmlab://resnet101_v1c'))))
diff --git a/configs/icnet/icnet_r101-d8_in1k-pre_832x832_80k_cityscapes.py b/configs/icnet/icnet_r101-d8_in1k-pre_832x832_80k_cityscapes.py
@@ -0,0 +1,12 @@
+_base_ = [
+    '../_base_/models/icnet_r50-d8.py',
+    '../_base_/datasets/cityscapes_832x832.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_80k.py'
+]
+
+model = dict(
+    backbone=dict(
+        backbone_cfg=dict(
+            depth=101,
+            init_cfg=dict(
+                type='Pretrained', checkpoint='open-mmlab://resnet101_v1c'))))