support coco stuff-10k/164k (#625)

* support coco stuff-10k/164k

* update docs

* fix docs

* update docs

* fix import lints

* Update docs/dataset_prepare.md

* Update docs/dataset_prepare.md

* Update tools/convert_datasets/coco_stuff164k.py

* Update tools/convert_datasets/coco_stuff10k.py

* Update tools/convert_datasets/coco_stuff10k.py

* Update tools/convert_datasets/coco_stuff10k.py

* Update tools/convert_datasets/coco_stuff10k.py

* Update coco_stuff.py

fix the description of the dataset

* Update dataset_prepare.md

fix the doc tree of coco stuff 10k

* Update coco_stuff10k.py

fix img_dir

* Update coco_stuff.py

fix descriptions

* Update coco_stuff164k.py

fix out_dir

* Update coco_stuff10k.py

fix save file name

* Update coco_stuff.py

fix seg_map_suffix

* Update dataset_prepare.md

fix -p

* Update dataset_prepare.md

fix doc tree

* modify coco stuff convertor

* Remove redundant code

* fix 164k convert bug

* remove redundant comment

* add deeplabv3 configs and more iterations

* replace shutil.move with shtil.copyfile

* Update deeplabv3_r50-d8_512x512_4x4_80k_coco_stuff10k.py

fix wrong config

* Update deeplabv3_r101-d8_512x512_4x4_80k_coco_stuff164k.py

fix wrong config

* fix wrong configs

* fix wrong configs

* fix wrong path for coco stuff 10k

* fix convert bugs

* fix seg_filename bug

* when nproc=0, use track progress

* rename configs: coco_stuff --> coco-stuff

* add coco-stuff 10k and 164k to README.md

* update configs

* add deeplabv3 benchmark

* add pspnet benchmark

* remove redundant comma

Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn>

README.md

@@ -102,12 +102,15 @@ Supported datasets:
 - [x] [PASCAL VOC](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#pascal-voc)
 - [x] [ADE20K](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#ade20k)
 - [x] [Pascal Context](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#pascal-context)
+- [x] [COCO-Stuff 10k](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#coco-stuff-10k)
+- [x] [COCO-Stuff 164k](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#coco-stuff-164k)
 - [x] [CHASE_DB1](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#chase-db1)
 - [x] [DRIVE](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#drive)
 - [x] [HRF](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#hrf)
 - [x] [STARE](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#stare)
 - [x] [Dark Zurich](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#dark-zurich)
 - [x] [Nighttime Driving](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#nighttime-driving)
+- [x] [Nighttime Driving](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#nighttime-driving)
 
 ## Installation
 

README_zh-CN.md

@@ -101,6 +101,8 @@ MMSegmentation 是一个基于 PyTorch 的语义分割开源工具箱。它是 O
 - [x] [PASCAL VOC](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#pascal-voc)
 - [x] [ADE20K](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#ade20k)
 - [x] [Pascal Context](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#pascal-context)
+- [x] [COCO-Stuff 10k](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#coco-stuff-10k)
+- [x] [COCO-Stuff 164k](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#coco-stuff-164k)
 - [x] [CHASE_DB1](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#chase-db1)
 - [x] [DRIVE](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#drive)
 - [x] [HRF](https://github.com/open-mmlab/mmsegmentation/blob/master/docs/dataset_prepare.md#hrf)

configs/_base_/datasets/coco-stuff10k.py

@@ -0,0 +1,57 @@
+# dataset settings
+dataset_type = 'COCOStuffDataset'
+data_root = 'data/coco_stuff10k'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+crop_size = (512, 512)
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', reduce_zero_label=True),
+    dict(type='Resize', img_scale=(2048, 512), 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, 512),
+        # 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(
+    samples_per_gpu=4,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        data_root=data_root,
+        reduce_zero_label=True,
+        img_dir='images/train2014',
+        ann_dir='annotations/train2014',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        data_root=data_root,
+        reduce_zero_label=True,
+        img_dir='images/test2014',
+        ann_dir='annotations/test2014',
+        pipeline=test_pipeline),
+    test=dict(
+        type=dataset_type,
+        data_root=data_root,
+        reduce_zero_label=True,
+        img_dir='images/test2014',
+        ann_dir='annotations/test2014',
+        pipeline=test_pipeline))

configs/_base_/datasets/coco-stuff164k.py

@@ -0,0 +1,54 @@
+# dataset settings
+dataset_type = 'COCOStuffDataset'
+data_root = 'data/coco_stuff164k'
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+crop_size = (512, 512)
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations'),
+    dict(type='Resize', img_scale=(2048, 512), 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, 512),
+        # 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(
+    samples_per_gpu=4,
+    workers_per_gpu=4,
+    train=dict(
+        type=dataset_type,
+        data_root=data_root,
+        img_dir='images/train2017',
+        ann_dir='annotations/train2017',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        data_root=data_root,
+        img_dir='images/val2017',
+        ann_dir='annotations/val2017',
+        pipeline=test_pipeline),
+    test=dict(
+        type=dataset_type,
+        data_root=data_root,
+        img_dir='images/val2017',
+        ann_dir='annotations/val2017',
+        pipeline=test_pipeline))

configs/_base_/schedules/schedule_320k.py

@@ -0,0 +1,9 @@
+# optimizer
+optimizer = dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0005)
+optimizer_config = dict()
+# learning policy
+lr_config = dict(policy='poly', power=0.9, min_lr=1e-4, by_epoch=False)
+# runtime settings
+runner = dict(type='IterBasedRunner', max_iters=320000)
+checkpoint_config = dict(by_epoch=False, interval=32000)
+evaluation = dict(interval=32000, metric='mIoU')

configs/deeplabv3/README.md

@@ -73,3 +73,23 @@
 | --------- | -------- | --------- | ------: | -------- | -------------- | ----: | ------------: | ------------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
 | DeepLabV3 | R-101-D8 | 480x480   |   40000 | -      | -           | 52.61 |         54.28 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3/deeplabv3_r101-d8_480x480_40k_pascal_context_59.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_480x480_40k_pascal_context_59/deeplabv3_r101-d8_480x480_40k_pascal_context_59_20210416_110332-cb08ea46.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_480x480_40k_pascal_context_59/deeplabv3_r101-d8_480x480_40k_pascal_context_59-20210416_110332.log.json) |
 | DeepLabV3 | R-101-D8 | 480x480   |   80000 | -        | -              | 52.46 |         54.09 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3/deeplabv3_r101-d8_480x480_80k_pascal_context_59.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_480x480_80k_pascal_context_59/deeplabv3_r101-d8_480x480_80k_pascal_context_59_20210416_113002-26303993.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_480x480_80k_pascal_context_59/deeplabv3_r101-d8_480x480_80k_pascal_context_59-20210416_113002.log.json) |
+
+### COCO-Stuff 10k
+
+| Method    | Backbone | Crop Size | Lr schd | Mem (GB) | Inf time (fps) |  mIoU | mIoU(ms+flip) | config                                                                                                                        | download                                                                                                                                                                                                                                                                                                                                                       |
+| --------- | -------- | --------- | ------: | -------- | -------------- | ----: | ------------: | ----------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| DeepLabV3 | R-50-D8  | 512x512   |   20000 | 9.6      | 10.8          | 34.66 |         36.08 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k.py)   | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k/deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k_20210821_043025-b35f789d.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k/deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k_20210821_043025.log.json)         |
+| DeepLabV3 | R-101-D8 | 512x512   |   20000 | 13.2     | 8.7          | 37.30 |         38.42 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3_r101-d8_512x512_4x4_20k_coco-stuff10k.py)  | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_20k_coco-stuff10k/deeplabv3_r101-d8_512x512_4x4_20k_coco-stuff10k_20210821_043025-c49752cb.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_20k_coco-stuff10k/deeplabv3_r101-d8_512x512_4x4_20k_coco-stuff10k_20210821_043025.log.json)     |
+| DeepLabV3 | R-50-D8  | 512x512   |  40000 | -        | -              | 35.73 |         37.09 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k.py)  | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k/deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k_20210821_043305-dc76f3ff.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k/deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k_20210821_043305.log.json)     |
+| DeepLabV3 | R-101-D8 | 512x512   |  40000 | -        | -              | 37.81 |         38.80 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3_r101-d8_512x512_4x4_40k_coco-stuff10k.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_40k_coco-stuff10k/deeplabv3_r101-d8_512x512_4x4_40k_coco-stuff10k_20210821_043305-636cb433.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_40k_coco-stuff10k/deeplabv3_r101-d8_512x512_4x4_40k_coco-stuff10k_20210821_043305.log.json) |
+
+### COCO-Stuff 164k
+
+| Method    | Backbone | Crop Size | Lr schd | Mem (GB) | Inf time (fps) |  mIoU | mIoU(ms+flip) | config                                                                                                                        | download                                                                                                                                                                                                                                                                                                                                                       |
+| --------- | -------- | --------- | ------: | -------- | -------------- | ----: | ------------: | ----------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| DeepLabV3 | R-50-D8  | 512x512   |   80000 | 9.6      | 10.8          | 39.38 |         40.03 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k.py)   | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k/deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k_20210709_163016-88675c24.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k/deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k_20210709_163016.log.json)         |
+| DeepLabV3 | R-101-D8 | 512x512   |   80000 | 13.2     | 8.7          | 40.87 |         41.50 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3_r101-d8_512x512_4x4_80k_coco-stuff164k.py)  | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_80k_coco-stuff164k/deeplabv3_r101-d8_512x512_4x4_80k_coco-stuff164k_20210709_201252-13600dc2.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_80k_coco-stuff164k/deeplabv3_r101-d8_512x512_4x4_80k_coco-stuff164k_20210709_201252.log.json)     |
+| DeepLabV3 | R-50-D8  | 512x512   |  160000 | -        | -              | 41.09 |         41.69 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k.py)  | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k/deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k_20210709_163016-49f2812b.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k/deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k_20210709_163016.log.json)     |
+| DeepLabV3 | R-101-D8 | 512x512   |  160000 | -        | -              | 41.82 |         42.49 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3_r101-d8_512x512_4x4_160k_coco-stuff164k.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_160k_coco-stuff164k/deeplabv3_r101-d8_512x512_4x4_160k_coco-stuff164k_20210709_155402-f035acfd.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_160k_coco-stuff164k/deeplabv3_r101-d8_512x512_4x4_160k_coco-stuff164k_20210709_155402.log.json) |
+| DeepLabV3 | R-50-D8  | 512x512   |  320000 | -        | -              | 41.37 |         42.22 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k.py)  | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k/deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k_20210709_155403-51b21115.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k/deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k_20210709_155403.log.json)     |
+| DeepLabV3 | R-101-D8 | 512x512   |  320000 | -        | -              | 42.61 |         43.42 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/deeplabv3_r101-d8_512x512_4x4_320k_coco-stuff164k.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_320k_coco-stuff164k/deeplabv3_r101-d8_512x512_4x4_320k_coco-stuff164k_20210709_155402-3cbca14d.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_320k_coco-stuff164k/deeplabv3_r101-d8_512x512_4x4_320k_coco-stuff164k_20210709_155402.log.json) |

configs/deeplabv3/deeplabv3.yml

@@ -6,6 +6,8 @@ Collections:
     - Pascal VOC 2012 + Aug
     - Pascal Context
     - Pascal Context 59
+    - COCO-Stuff 10k
+    - COCO-Stuff 164k
   Name: deeplabv3
 Models:
 - Config: configs/deeplabv3/deeplabv3_r50-d8_512x1024_40k_cityscapes.py
@@ -550,3 +552,175 @@ Models:
       mIoU(ms+flip): 54.09
     Task: Semantic Segmentation
   Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_480x480_80k_pascal_context_59/deeplabv3_r101-d8_480x480_80k_pascal_context_59_20210416_113002-26303993.pth
+- Config: configs/deeplabv3/deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k.py
+  In Collection: deeplabv3
+  Metadata:
+    backbone: R-50-D8
+    crop size: (512,512)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (512,512)
+      value: 92.59
+    lr schd: 20000
+    memory (GB): 9.6
+  Name: deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k
+  Results:
+    Dataset: COCO-Stuff 10k
+    Metrics:
+      mIoU: 34.66
+      mIoU(ms+flip): 36.08
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k/deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k_20210821_043025-b35f789d.pth
+- Config: configs/deeplabv3/deeplabv3_r101-d8_512x512_4x4_20k_coco-stuff10k.py
+  In Collection: deeplabv3
+  Metadata:
+    backbone: R-101-D8
+    crop size: (512,512)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (512,512)
+      value: 114.94
+    lr schd: 20000
+    memory (GB): 13.2
+  Name: deeplabv3_r101-d8_512x512_4x4_20k_coco-stuff10k
+  Results:
+    Dataset: COCO-Stuff 10k
+    Metrics:
+      mIoU: 37.3
+      mIoU(ms+flip): 38.42
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_20k_coco-stuff10k/deeplabv3_r101-d8_512x512_4x4_20k_coco-stuff10k_20210821_043025-c49752cb.pth
+- Config: configs/deeplabv3/deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k.py
+  In Collection: deeplabv3
+  Metadata:
+    backbone: R-50-D8
+    crop size: (512,512)
+    lr schd: 40000
+  Name: deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k
+  Results:
+    Dataset: COCO-Stuff 10k
+    Metrics:
+      mIoU: 35.73
+      mIoU(ms+flip): 37.09
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k/deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k_20210821_043305-dc76f3ff.pth
+- Config: configs/deeplabv3/deeplabv3_r101-d8_512x512_4x4_40k_coco-stuff10k.py
+  In Collection: deeplabv3
+  Metadata:
+    backbone: R-101-D8
+    crop size: (512,512)
+    lr schd: 40000
+  Name: deeplabv3_r101-d8_512x512_4x4_40k_coco-stuff10k
+  Results:
+    Dataset: COCO-Stuff 10k
+    Metrics:
+      mIoU: 37.81
+      mIoU(ms+flip): 38.8
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_40k_coco-stuff10k/deeplabv3_r101-d8_512x512_4x4_40k_coco-stuff10k_20210821_043305-636cb433.pth
+- Config: configs/deeplabv3/deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k.py
+  In Collection: deeplabv3
+  Metadata:
+    backbone: R-50-D8
+    crop size: (512,512)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (512,512)
+      value: 92.59
+    lr schd: 80000
+    memory (GB): 9.6
+  Name: deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 39.38
+      mIoU(ms+flip): 40.03
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k/deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k_20210709_163016-88675c24.pth
+- Config: configs/deeplabv3/deeplabv3_r101-d8_512x512_4x4_80k_coco-stuff164k.py
+  In Collection: deeplabv3
+  Metadata:
+    backbone: R-101-D8
+    crop size: (512,512)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (512,512)
+      value: 114.94
+    lr schd: 80000
+    memory (GB): 13.2
+  Name: deeplabv3_r101-d8_512x512_4x4_80k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 40.87
+      mIoU(ms+flip): 41.5
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_80k_coco-stuff164k/deeplabv3_r101-d8_512x512_4x4_80k_coco-stuff164k_20210709_201252-13600dc2.pth
+- Config: configs/deeplabv3/deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k.py
+  In Collection: deeplabv3
+  Metadata:
+    backbone: R-50-D8
+    crop size: (512,512)
+    lr schd: 160000
+  Name: deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 41.09
+      mIoU(ms+flip): 41.69
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k/deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k_20210709_163016-49f2812b.pth
+- Config: configs/deeplabv3/deeplabv3_r101-d8_512x512_4x4_160k_coco-stuff164k.py
+  In Collection: deeplabv3
+  Metadata:
+    backbone: R-101-D8
+    crop size: (512,512)
+    lr schd: 160000
+  Name: deeplabv3_r101-d8_512x512_4x4_160k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 41.82
+      mIoU(ms+flip): 42.49
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_160k_coco-stuff164k/deeplabv3_r101-d8_512x512_4x4_160k_coco-stuff164k_20210709_155402-f035acfd.pth
+- Config: configs/deeplabv3/deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k.py
+  In Collection: deeplabv3
+  Metadata:
+    backbone: R-50-D8
+    crop size: (512,512)
+    lr schd: 320000
+  Name: deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 41.37
+      mIoU(ms+flip): 42.22
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k/deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k_20210709_155403-51b21115.pth
+- Config: configs/deeplabv3/deeplabv3_r101-d8_512x512_4x4_320k_coco-stuff164k.py
+  In Collection: deeplabv3
+  Metadata:
+    backbone: R-101-D8
+    crop size: (512,512)
+    lr schd: 320000
+  Name: deeplabv3_r101-d8_512x512_4x4_320k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 42.61
+      mIoU(ms+flip): 43.42
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/deeplabv3/deeplabv3_r101-d8_512x512_4x4_320k_coco-stuff164k/deeplabv3_r101-d8_512x512_4x4_320k_coco-stuff164k_20210709_155402-3cbca14d.pth

configs/deeplabv3/deeplabv3_r101-d8_512x512_4x4_160k_coco-stuff164k.py

@@ -0,0 +1,2 @@
+_base_ = './deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k.py'
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

configs/deeplabv3/deeplabv3_r101-d8_512x512_4x4_20k_coco-stuff10k.py

@@ -0,0 +1,2 @@
+_base_ = './deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k.py'
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

configs/deeplabv3/deeplabv3_r101-d8_512x512_4x4_320k_coco-stuff164k.py

@@ -0,0 +1,2 @@
+_base_ = './deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k.py'
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

configs/deeplabv3/deeplabv3_r101-d8_512x512_4x4_40k_coco-stuff10k.py

@@ -0,0 +1,2 @@
+_base_ = './deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k.py'
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

configs/deeplabv3/deeplabv3_r101-d8_512x512_4x4_80k_coco-stuff164k.py

@@ -0,0 +1,2 @@
+_base_ = './deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k.py'
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

configs/deeplabv3/deeplabv3_r50-d8_512x512_4x4_160k_coco-stuff164k.py

@@ -0,0 +1,7 @@
+_base_ = [
+    '../_base_/models/deeplabv3_r50-d8.py',
+    '../_base_/datasets/coco-stuff164k.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_160k.py'
+]
+model = dict(
+    decode_head=dict(num_classes=171), auxiliary_head=dict(num_classes=171))

configs/deeplabv3/deeplabv3_r50-d8_512x512_4x4_20k_coco-stuff10k.py

@@ -0,0 +1,7 @@
+_base_ = [
+    '../_base_/models/deeplabv3_r50-d8.py',
+    '../_base_/datasets/coco-stuff10k.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_20k.py'
+]
+model = dict(
+    decode_head=dict(num_classes=171), auxiliary_head=dict(num_classes=171))

configs/deeplabv3/deeplabv3_r50-d8_512x512_4x4_320k_coco-stuff164k.py

@@ -0,0 +1,7 @@
+_base_ = [
+    '../_base_/models/deeplabv3_r50-d8.py',
+    '../_base_/datasets/coco-stuff164k.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_320k.py'
+]
+model = dict(
+    decode_head=dict(num_classes=171), auxiliary_head=dict(num_classes=171))

configs/deeplabv3/deeplabv3_r50-d8_512x512_4x4_40k_coco-stuff10k.py

@@ -0,0 +1,7 @@
+_base_ = [
+    '../_base_/models/deeplabv3_r50-d8.py',
+    '../_base_/datasets/coco-stuff10k.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_40k.py'
+]
+model = dict(
+    decode_head=dict(num_classes=171), auxiliary_head=dict(num_classes=171))

configs/deeplabv3/deeplabv3_r50-d8_512x512_4x4_80k_coco-stuff164k.py

@@ -0,0 +1,7 @@
+_base_ = [
+    '../_base_/models/deeplabv3_r50-d8.py',
+    '../_base_/datasets/coco-stuff164k.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_80k.py'
+]
+model = dict(
+    decode_head=dict(num_classes=171), auxiliary_head=dict(num_classes=171))

configs/pspnet/README.md

@@ -83,3 +83,23 @@ We support evaluation results on these two datasets using models above trained o
   |PSPNet|R-101b-D8|Cityscapes Training set |Dark Zurich      |15.54|[config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet/pspnet_r101b-d8_512x1024_80k_dark.py)| [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101b-d8_512x1024_80k_cityscapes/pspnet_r101b-d8_512x1024_80k_cityscapes_20201226_170012-3a4d38ab.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101b-d8_512x1024_80k_cityscapes/pspnet_r101b-d8_512x1024_80k_cityscapes-20201226_170012.log.json) |
   |PSPNet|R-101b-D8|Cityscapes Training set |Nighttime Driving|22.25|[config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet/pspnet_r101b-d8_512x1024_80k_night_driving.py)| [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101b-d8_512x1024_80k_cityscapes/pspnet_r101b-d8_512x1024_80k_cityscapes_20201226_170012-3a4d38ab.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101b-d8_512x1024_80k_cityscapes/pspnet_r101b-d8_512x1024_80k_cityscapes-20201226_170012.log.json) |
   |PSPNet|R-101b-D8|Cityscapes Training set |Cityscapes Validation set|79.69|[config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet/pspnet_r101b-d8_512x1024_80k_cityscapes.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101b-d8_512x1024_80k_cityscapes/pspnet_r101b-d8_512x1024_80k_cityscapes_20201226_170012-3a4d38ab.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101b-d8_512x1024_80k_cityscapes/pspnet_r101b-d8_512x1024_80k_cityscapes-20201226_170012.log.json) |
+
+### COCO-Stuff 10k
+
+| Method    | Backbone | Crop Size | Lr schd | Mem (GB) | Inf time (fps) |  mIoU | mIoU(ms+flip) | config                                                                                                                        | download                                                                                                                                                                                                                                                                                                                                                       |
+| --------- | -------- | --------- | ------: | -------- | -------------- | ----: | ------------: | ----------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| PSPNet | R-50-D8  | 512x512   |   20000 | 9.6      | 20.5          | 35.69 |         36.62 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k.py)   | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k/pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k_20210820_203258-b88df27f.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k/pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k_20210820_203258.log.json)         |
+| PSPNet | R-101-D8 | 512x512   |   20000 | 13.2     | 11.1          | 37.26 |         38.52 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet_r101-d8_512x512_4x4_20k_coco-stuff10k.py)  | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_20k_coco-stuff10k/pspnet_r101-d8_512x512_4x4_20k_coco-stuff10k_20210820_232135-76aae482.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_20k_coco-stuff10k/pspnet_r101-d8_512x512_4x4_20k_coco-stuff10k_20210820_232135.log.json)     |
+| PSPNet | R-50-D8  | 512x512   |  40000 | -        | -              | 36.33 |         37.24 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k.py)  | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k/pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k_20210821_030857-92e2902b.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k/pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k_20210821_030857.log.json)     |
+| PSPNet | R-101-D8 | 512x512   |  40000 | -        | -              | 37.76 |         38.86 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet_r101-d8_512x512_4x4_40k_coco-stuff10k.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_40k_coco-stuff10k/pspnet_r101-d8_512x512_4x4_40k_coco-stuff10k_20210821_014022-831aec95.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_40k_coco-stuff10k/pspnet_r101-d8_512x512_4x4_40k_coco-stuff10k_20210821_014022.log.json) |
+
+### COCO-Stuff 164k
+
+| Method    | Backbone | Crop Size | Lr schd | Mem (GB) | Inf time (fps) |  mIoU | mIoU(ms+flip) | config                                                                                                                        | download                                                                                                                                                                                                                                                                                                                                                       |
+| --------- | -------- | --------- | ------: | -------- | -------------- | ----: | ------------: | ----------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| PSPNet | R-50-D8  | 512x512   |   80000 | 9.6      | 20.5          | 38.80 |         39.19 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k.py)   | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k/pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k_20210707_152034-0e41b2db.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k/pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k_20210707_152034.log.json)         |
+| PSPNet | R-101-D8 | 512x512   |   80000 | 13.2     | 11.1          | 40.34 |         40.79 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet_r101-d8_512x512_4x4_80k_coco-stuff164k.py)  | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_80k_coco-stuff164k/pspnet_r101-d8_512x512_4x4_80k_coco-stuff164k_20210707_152034-7eb41789.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_80k_coco-stuff164k/pspnet_r101-d8_512x512_4x4_80k_coco-stuff164k_20210707_152034.log.json)     |
+| PSPNet | R-50-D8  | 512x512   |  160000 | -        | -              | 39.64 |         39.97 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k.py)  | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k/pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k_20210707_152004-51276a57.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k/pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k_20210707_152004.log.json)     |
+| PSPNet | R-101-D8 | 512x512   |  160000 | -        | -              | 41.28 |         41.66 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet_r101-d8_512x512_4x4_160k_coco-stuff164k.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_160k_coco-stuff164k/pspnet_r101-d8_512x512_4x4_160k_coco-stuff164k_20210707_152004-4af9621b.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_160k_coco-stuff164k/pspnet_r101-d8_512x512_4x4_160k_coco-stuff164k_20210707_152004.log.json) |
+| PSPNet | R-50-D8  | 512x512   |  320000 | -        | -              | 40.53 |         40.75 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k.py)  | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k/pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k_20210707_152004-be9610cc.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k/pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k_20210707_152004.log.json)     |
+| PSPNet | R-101-D8 | 512x512   |  320000 | -        | -              | 41.95 |         42.42 | [config](https://github.com/open-mmlab/mmsegmentation/blob/master/configs/pspnet_r101-d8_512x512_4x4_320k_coco-stuff164k.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_320k_coco-stuff164k/pspnet_r101-d8_512x512_4x4_320k_coco-stuff164k_20210707_152004-72220c60.pth) | [log](https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_320k_coco-stuff164k/pspnet_r101-d8_512x512_4x4_320k_coco-stuff164k_20210707_152004.log.json) |

configs/pspnet/pspnet.yml

@@ -7,6 +7,8 @@ Collections:
     - Pascal Context
     - Pascal Context 59
     - Dark Zurich and Nighttime Driving
+    - COCO-Stuff 10k
+    - COCO-Stuff 164k
   Name: pspnet
 Models:
 - Config: configs/pspnet/pspnet_r50-d8_512x1024_40k_cityscapes.py
@@ -537,3 +539,175 @@ Models:
       mIoU(ms+flip): 53.99
     Task: Semantic Segmentation
   Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_480x480_80k_pascal_context_59/pspnet_r101-d8_480x480_80k_pascal_context_59_20210416_114418-fa6caaa2.pth
+- Config: configs/pspnet/pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k.py
+  In Collection: pspnet
+  Metadata:
+    backbone: R-50-D8
+    crop size: (512,512)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (512,512)
+      value: 48.78
+    lr schd: 20000
+    memory (GB): 9.6
+  Name: pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k
+  Results:
+    Dataset: COCO-Stuff 10k
+    Metrics:
+      mIoU: 35.69
+      mIoU(ms+flip): 36.62
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k/pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k_20210820_203258-b88df27f.pth
+- Config: configs/pspnet/pspnet_r101-d8_512x512_4x4_20k_coco-stuff10k.py
+  In Collection: pspnet
+  Metadata:
+    backbone: R-101-D8
+    crop size: (512,512)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (512,512)
+      value: 90.09
+    lr schd: 20000
+    memory (GB): 13.2
+  Name: pspnet_r101-d8_512x512_4x4_20k_coco-stuff10k
+  Results:
+    Dataset: COCO-Stuff 10k
+    Metrics:
+      mIoU: 37.26
+      mIoU(ms+flip): 38.52
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_20k_coco-stuff10k/pspnet_r101-d8_512x512_4x4_20k_coco-stuff10k_20210820_232135-76aae482.pth
+- Config: configs/pspnet/pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k.py
+  In Collection: pspnet
+  Metadata:
+    backbone: R-50-D8
+    crop size: (512,512)
+    lr schd: 40000
+  Name: pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k
+  Results:
+    Dataset: COCO-Stuff 10k
+    Metrics:
+      mIoU: 36.33
+      mIoU(ms+flip): 37.24
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k/pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k_20210821_030857-92e2902b.pth
+- Config: configs/pspnet/pspnet_r101-d8_512x512_4x4_40k_coco-stuff10k.py
+  In Collection: pspnet
+  Metadata:
+    backbone: R-101-D8
+    crop size: (512,512)
+    lr schd: 40000
+  Name: pspnet_r101-d8_512x512_4x4_40k_coco-stuff10k
+  Results:
+    Dataset: COCO-Stuff 10k
+    Metrics:
+      mIoU: 37.76
+      mIoU(ms+flip): 38.86
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_40k_coco-stuff10k/pspnet_r101-d8_512x512_4x4_40k_coco-stuff10k_20210821_014022-831aec95.pth
+- Config: configs/pspnet/pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k.py
+  In Collection: pspnet
+  Metadata:
+    backbone: R-50-D8
+    crop size: (512,512)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (512,512)
+      value: 48.78
+    lr schd: 80000
+    memory (GB): 9.6
+  Name: pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 38.8
+      mIoU(ms+flip): 39.19
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k/pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k_20210707_152034-0e41b2db.pth
+- Config: configs/pspnet/pspnet_r101-d8_512x512_4x4_80k_coco-stuff164k.py
+  In Collection: pspnet
+  Metadata:
+    backbone: R-101-D8
+    crop size: (512,512)
+    inference time (ms/im):
+    - backend: PyTorch
+      batch size: 1
+      hardware: V100
+      mode: FP32
+      resolution: (512,512)
+      value: 90.09
+    lr schd: 80000
+    memory (GB): 13.2
+  Name: pspnet_r101-d8_512x512_4x4_80k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 40.34
+      mIoU(ms+flip): 40.79
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_80k_coco-stuff164k/pspnet_r101-d8_512x512_4x4_80k_coco-stuff164k_20210707_152034-7eb41789.pth
+- Config: configs/pspnet/pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k.py
+  In Collection: pspnet
+  Metadata:
+    backbone: R-50-D8
+    crop size: (512,512)
+    lr schd: 160000
+  Name: pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 39.64
+      mIoU(ms+flip): 39.97
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k/pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k_20210707_152004-51276a57.pth
+- Config: configs/pspnet/pspnet_r101-d8_512x512_4x4_160k_coco-stuff164k.py
+  In Collection: pspnet
+  Metadata:
+    backbone: R-101-D8
+    crop size: (512,512)
+    lr schd: 160000
+  Name: pspnet_r101-d8_512x512_4x4_160k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 41.28
+      mIoU(ms+flip): 41.66
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_160k_coco-stuff164k/pspnet_r101-d8_512x512_4x4_160k_coco-stuff164k_20210707_152004-4af9621b.pth
+- Config: configs/pspnet/pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k.py
+  In Collection: pspnet
+  Metadata:
+    backbone: R-50-D8
+    crop size: (512,512)
+    lr schd: 320000
+  Name: pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 40.53
+      mIoU(ms+flip): 40.75
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k/pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k_20210707_152004-be9610cc.pth
+- Config: configs/pspnet/pspnet_r101-d8_512x512_4x4_320k_coco-stuff164k.py
+  In Collection: pspnet
+  Metadata:
+    backbone: R-101-D8
+    crop size: (512,512)
+    lr schd: 320000
+  Name: pspnet_r101-d8_512x512_4x4_320k_coco-stuff164k
+  Results:
+    Dataset: COCO-Stuff 164k
+    Metrics:
+      mIoU: 41.95
+      mIoU(ms+flip): 42.42
+    Task: Semantic Segmentation
+  Weights: https://download.openmmlab.com/mmsegmentation/v0.5/pspnet/pspnet_r101-d8_512x512_4x4_320k_coco-stuff164k/pspnet_r101-d8_512x512_4x4_320k_coco-stuff164k_20210707_152004-72220c60.pth

configs/pspnet/pspnet_r101-d8_512x512_4x4_160k_coco-stuff164k.py

@@ -0,0 +1,2 @@
+_base_ = './pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k.py'
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

configs/pspnet/pspnet_r101-d8_512x512_4x4_20k_coco-stuff10k.py

@@ -0,0 +1,2 @@
+_base_ = './pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k.py'
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

configs/pspnet/pspnet_r101-d8_512x512_4x4_320k_coco-stuff164k.py

@@ -0,0 +1,2 @@
+_base_ = './pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k.py'
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

configs/pspnet/pspnet_r101-d8_512x512_4x4_40k_coco-stuff10k.py

@@ -0,0 +1,2 @@
+_base_ = './pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k.py'
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

configs/pspnet/pspnet_r101-d8_512x512_4x4_80k_coco-stuff164k.py

@@ -0,0 +1,2 @@
+_base_ = './pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k.py'
+model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

configs/pspnet/pspnet_r50-d8_512x512_4x4_160k_coco-stuff164k.py

@@ -0,0 +1,7 @@
+_base_ = [
+    '../_base_/models/pspnet_r50-d8.py',
+    '../_base_/datasets/coco-stuff164k.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_160k.py'
+]
+model = dict(
+    decode_head=dict(num_classes=171), auxiliary_head=dict(num_classes=171))

configs/pspnet/pspnet_r50-d8_512x512_4x4_20k_coco-stuff10k.py

@@ -0,0 +1,6 @@
+_base_ = [
+    '../_base_/models/pspnet_r50-d8.py', '../_base_/datasets/coco-stuff10k.py',
+    '../_base_/default_runtime.py', '../_base_/schedules/schedule_20k.py'
+]
+model = dict(
+    decode_head=dict(num_classes=171), auxiliary_head=dict(num_classes=171))

configs/pspnet/pspnet_r50-d8_512x512_4x4_320k_coco-stuff164k.py

@@ -0,0 +1,7 @@
+_base_ = [
+    '../_base_/models/pspnet_r50-d8.py',
+    '../_base_/datasets/coco-stuff164k.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_320k.py'
+]
+model = dict(
+    decode_head=dict(num_classes=171), auxiliary_head=dict(num_classes=171))

configs/pspnet/pspnet_r50-d8_512x512_4x4_40k_coco-stuff10k.py

@@ -0,0 +1,6 @@
+_base_ = [
+    '../_base_/models/pspnet_r50-d8.py', '../_base_/datasets/coco-stuff10k.py',
+    '../_base_/default_runtime.py', '../_base_/schedules/schedule_40k.py'
+]
+model = dict(
+    decode_head=dict(num_classes=171), auxiliary_head=dict(num_classes=171))

configs/pspnet/pspnet_r50-d8_512x512_4x4_80k_coco-stuff164k.py

@@ -0,0 +1,7 @@
+_base_ = [
+    '../_base_/models/pspnet_r50-d8.py',
+    '../_base_/datasets/coco-stuff164k.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_80k.py'
+]
+model = dict(
+    decode_head=dict(num_classes=171), auxiliary_head=dict(num_classes=171))

docs/dataset_prepare.md

@@ -41,6 +41,24 @@ mmsegmentation
 │   │   │   ├── images
 │   │   │   │   ├── training
 │   │   │   │   ├── validation
+│   ├── coco_stuff10k
+│   │   ├── images
+│   │   │   ├── train2014
+│   │   │   ├── test2014
+│   │   ├── annotations
+│   │   │   ├── train2014
+│   │   │   ├── test2014
+│   │   ├── imagesLists
+│   │   │   ├── train.txt
+│   │   │   ├── test.txt
+│   │   │   ├── all.txt
+│   ├── coco_stuff164k
+│   │   ├── images
+│   │   │   ├── train2017
+│   │   │   ├── val2017
+│   │   ├── annotations
+│   │   │   ├── train2017
+│   │   │   ├── val2017
 │   ├── CHASE_DB1
 │   │   ├── images
 │   │   │   ├── training
@@ -136,6 +154,50 @@ If you would like to use Pascal Context dataset, please install [Detail](https:/
 python tools/convert_datasets/pascal_context.py data/VOCdevkit data/VOCdevkit/VOC2010/trainval_merged.json
 ```
 
+### COCO Stuff 10k
+
+The data could be downloaded [here](http://calvin.inf.ed.ac.uk/wp-content/uploads/data/cocostuffdataset/cocostuff-10k-v1.1.zip) by wget.
+
+For COCO Stuff 10k dataset, please run the following commands to download and convert the dataset.
+
+```shell
+# download
+mkdir coco_stuff10k && cd coco_stuff10k
+wget http://calvin.inf.ed.ac.uk/wp-content/uploads/data/cocostuffdataset/cocostuff-10k-v1.1.zip
+
+# unzip
+unzip cocostuff-10k-v1.1.zip
+
+# --nproc means 8 process for conversion, which could be omitted as well.
+python tools/convert_datasets/coco_stuff10k.py /path/to/coco_stuff10k --nproc 8
+```
+
+By convention, mask labels in `/path/to/coco_stuff164k/annotations/*2014/*_labelTrainIds.png` are used for COCO Stuff 10k training and testing.
+
+### COCO Stuff 164k
+
+For COCO Stuff 164k dataset, please run the following commands to download and convert the augmented dataset.
+
+```shell
+# download
+mkdir coco_stuff164k && cd coco_stuff164k
+wget http://images.cocodataset.org/zips/train2017.zip
+wget http://images.cocodataset.org/zips/val2017.zip
+wget http://calvin.inf.ed.ac.uk/wp-content/uploads/data/cocostuffdataset/stuffthingmaps_trainval2017.zip
+
+# unzip
+unzip train2017.zip -d images/
+unzip val2017.zip -d images/
+unzip stuffthingmaps_trainval2017.zip -d annotations/
+
+# --nproc means 8 process for conversion, which could be omitted as well.
+python tools/convert_datasets/coco_stuff164k.py /path/to/coco_stuff164k --nproc 8
+```
+
+By convention, mask labels in `/path/to/coco_stuff164k/annotations/*2017/*_labelTrainIds.png` are used for COCO Stuff 164k training and testing.
+
+The details of this dataset could be found at [here](https://github.com/nightrome/cocostuff#downloads).
+
 ### CHASE DB1
 
 The training and validation set of CHASE DB1 could be download from [here](https://staffnet.kingston.ac.uk/~ku15565/CHASE_DB1/assets/CHASEDB1.zip).

mmseg/datasets/__init__.py

@@ -3,6 +3,7 @@ from .ade import ADE20KDataset
 from .builder import DATASETS, PIPELINES, build_dataloader, build_dataset
 from .chase_db1 import ChaseDB1Dataset
 from .cityscapes import CityscapesDataset
+from .coco_stuff import COCOStuffDataset
 from .custom import CustomDataset
 from .dark_zurich import DarkZurichDataset
 from .dataset_wrappers import ConcatDataset, RepeatDataset
@@ -18,5 +19,6 @@ __all__ = [
     'DATASETS', 'build_dataset', 'PIPELINES', 'CityscapesDataset',
     'PascalVOCDataset', 'ADE20KDataset', 'PascalContextDataset',
     'PascalContextDataset59', 'ChaseDB1Dataset', 'DRIVEDataset', 'HRFDataset',
-    'STAREDataset', 'DarkZurichDataset', 'NightDrivingDataset'
+    'STAREDataset', 'DarkZurichDataset', 'NightDrivingDataset',
+    'COCOStuffDataset'
 ]

mmseg/datasets/coco_stuff.py

@@ -0,0 +1,93 @@
+from .builder import DATASETS
+from .custom import CustomDataset
+
+
+@DATASETS.register_module()
+class COCOStuffDataset(CustomDataset):
+    """COCO-Stuff dataset.
+
+    In segmentation map annotation for COCO-Stuff, Train-IDs of the 10k version
+    are from 1 to 171, where 0 is the ignore index, and Train-ID of COCO Stuff
+    164k is from 0 to 170, where 255 is the ignore index. So, they are all 171
+    semantic categories. ``reduce_zero_label`` is set to True and False for the
+    10k and 164k versions, respectively. The ``img_suffix`` is fixed to '.jpg',
+    and ``seg_map_suffix`` is fixed to '.png'.
+    """
+    CLASSES = (
+        'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train',
+        'truck', 'boat', 'traffic light', 'fire hydrant', 'stop sign',
+        'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep',
+        'cow', 'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella',
+        'handbag', 'tie', 'suitcase', 'frisbee', 'skis', 'snowboard',
+        'sports ball', 'kite', 'baseball bat', 'baseball glove', 'skateboard',
+        'surfboard', 'tennis racket', 'bottle', 'wine glass', 'cup', 'fork',
+        'knife', 'spoon', 'bowl', 'banana', 'apple', 'sandwich', 'orange',
+        'broccoli', 'carrot', 'hot dog', 'pizza', 'donut', 'cake', 'chair',
+        'couch', 'potted plant', 'bed', 'dining table', 'toilet', 'tv',
+        'laptop', 'mouse', 'remote', 'keyboard', 'cell phone', 'microwave',
+        'oven', 'toaster', 'sink', 'refrigerator', 'book', 'clock', 'vase',
+        'scissors', 'teddy bear', 'hair drier', 'toothbrush', 'banner',
+        'blanket', 'branch', 'bridge', 'building-other', 'bush', 'cabinet',
+        'cage', 'cardboard', 'carpet', 'ceiling-other', 'ceiling-tile',
+        'cloth', 'clothes', 'clouds', 'counter', 'cupboard', 'curtain',
+        'desk-stuff', 'dirt', 'door-stuff', 'fence', 'floor-marble',
+        'floor-other', 'floor-stone', 'floor-tile', 'floor-wood',
+        'flower', 'fog', 'food-other', 'fruit', 'furniture-other', 'grass',
+        'gravel', 'ground-other', 'hill', 'house', 'leaves', 'light', 'mat',
+        'metal', 'mirror-stuff', 'moss', 'mountain', 'mud', 'napkin', 'net',
+        'paper', 'pavement', 'pillow', 'plant-other', 'plastic', 'platform',
+        'playingfield', 'railing', 'railroad', 'river', 'road', 'rock', 'roof',
+        'rug', 'salad', 'sand', 'sea', 'shelf', 'sky-other', 'skyscraper',
+        'snow', 'solid-other', 'stairs', 'stone', 'straw', 'structural-other',
+        'table', 'tent', 'textile-other', 'towel', 'tree', 'vegetable',
+        'wall-brick', 'wall-concrete', 'wall-other', 'wall-panel',
+        'wall-stone', 'wall-tile', 'wall-wood', 'water-other', 'waterdrops',
+        'window-blind', 'window-other', 'wood')
+
+    PALETTE = [[0, 192, 64], [0, 192, 64], [0, 64, 96], [128, 192, 192],
+               [0, 64, 64], [0, 192, 224], [0, 192, 192], [128, 192, 64],
+               [0, 192, 96], [128, 192, 64], [128, 32, 192], [0, 0, 224],
+               [0, 0, 64], [0, 160, 192], [128, 0, 96], [128, 0, 192],
+               [0, 32, 192], [128, 128, 224], [0, 0, 192], [128, 160, 192],
+               [128, 128, 0], [128, 0, 32], [128, 32, 0], [128, 0, 128],
+               [64, 128, 32], [0, 160, 0], [0, 0, 0], [192, 128, 160],
+               [0, 32, 0], [0, 128, 128], [64, 128, 160], [128, 160, 0],
+               [0, 128, 0], [192, 128, 32], [128, 96, 128], [0, 0, 128],
+               [64, 0, 32], [0, 224, 128], [128, 0, 0], [192, 0, 160],
+               [0, 96, 128], [128, 128, 128], [64, 0, 160], [128, 224, 128],
+               [128, 128, 64], [192, 0, 32], [128, 96, 0], [128, 0, 192],
+               [0, 128, 32], [64, 224, 0], [0, 0, 64], [128, 128, 160],
+               [64, 96, 0], [0, 128, 192], [0, 128, 160], [192, 224, 0],
+               [0, 128, 64], [128, 128, 32], [192, 32, 128], [0, 64, 192],
+               [0, 0, 32], [64, 160, 128], [128, 64, 64], [128, 0, 160],
+               [64, 32, 128], [128, 192, 192], [0, 0, 160], [192, 160, 128],
+               [128, 192, 0], [128, 0, 96], [192, 32, 0], [128, 64, 128],
+               [64, 128, 96], [64, 160, 0], [0, 64, 0], [192, 128, 224],
+               [64, 32, 0], [0, 192, 128], [64, 128, 224], [192, 160, 0],
+               [0, 192, 0], [192, 128, 96], [192, 96, 128], [0, 64, 128],
+               [64, 0, 96], [64, 224, 128], [128, 64, 0], [192, 0, 224],
+               [64, 96, 128], [128, 192, 128], [64, 0, 224], [192, 224, 128],
+               [128, 192, 64], [192, 0, 96], [192, 96, 0], [128, 64, 192],
+               [0, 128, 96], [0, 224, 0], [64, 64, 64], [128, 128, 224],
+               [0, 96, 0], [64, 192, 192], [0, 128, 224], [128, 224, 0],
+               [64, 192, 64], [128, 128, 96], [128, 32, 128], [64, 0, 192],
+               [0, 64, 96], [0, 160, 128], [192, 0, 64], [128, 64, 224],
+               [0, 32, 128], [192, 128, 192], [0, 64, 224], [128, 160, 128],
+               [192, 128, 0], [128, 64, 32], [128, 32, 64], [192, 0, 128],
+               [64, 192, 32], [0, 160, 64], [64, 0, 0], [192, 192, 160],
+               [0, 32, 64], [64, 128, 128], [64, 192, 160], [128, 160, 64],
+               [64, 128, 0], [192, 192, 32], [128, 96, 192], [64, 0, 128],
+               [64, 64, 32], [0, 224, 192], [192, 0, 0], [192, 64, 160],
+               [0, 96, 192], [192, 128, 128], [64, 64, 160], [128, 224, 192],
+               [192, 128, 64], [192, 64, 32], [128, 96, 64], [192, 0, 192],
+               [0, 192, 32], [64, 224, 64], [64, 0, 64], [128, 192, 160],
+               [64, 96, 64], [64, 128, 192], [0, 192, 160], [192, 224, 64],
+               [64, 128, 64], [128, 192, 32], [192, 32, 192], [64, 64, 192],
+               [0, 64, 32], [64, 160, 192], [192, 64, 64], [128, 64, 160],
+               [64, 32, 192], [192, 192, 192], [0, 64, 160], [192, 160, 192],
+               [192, 192, 0], [128, 64, 96], [192, 32, 64], [192, 64, 128],
+               [64, 192, 96], [64, 160, 64], [64, 64, 0]]
+
+    def __init__(self, **kwargs):
+        super(COCOStuffDataset, self).__init__(
+            img_suffix='.jpg', seg_map_suffix='_labelTrainIds.png', **kwargs)

tools/convert_datasets/coco_stuff10k.py

@@ -0,0 +1,306 @@
+import argparse
+import os.path as osp
+import shutil
+from functools import partial
+
+import mmcv
+import numpy as np
+from PIL import Image
+from scipy.io import loadmat
+
+COCO_LEN = 10000
+
+clsID_to_trID = {
+    0: 0,
+    1: 1,
+    2: 2,
+    3: 3,
+    4: 4,
+    5: 5,
+    6: 6,
+    7: 7,
+    8: 8,
+    9: 9,
+    10: 10,
+    11: 11,
+    13: 12,
+    14: 13,
+    15: 14,
+    16: 15,
+    17: 16,
+    18: 17,
+    19: 18,
+    20: 19,
+    21: 20,
+    22: 21,
+    23: 22,
+    24: 23,
+    25: 24,
+    27: 25,
+    28: 26,
+    31: 27,
+    32: 28,
+    33: 29,
+    34: 30,
+    35: 31,
+    36: 32,
+    37: 33,
+    38: 34,
+    39: 35,
+    40: 36,
+    41: 37,
+    42: 38,
+    43: 39,
+    44: 40,
+    46: 41,
+    47: 42,
+    48: 43,
+    49: 44,
+    50: 45,
+    51: 46,
+    52: 47,
+    53: 48,
+    54: 49,
+    55: 50,
+    56: 51,
+    57: 52,
+    58: 53,
+    59: 54,
+    60: 55,
+    61: 56,
+    62: 57,
+    63: 58,
+    64: 59,
+    65: 60,
+    67: 61,
+    70: 62,
+    72: 63,
+    73: 64,
+    74: 65,
+    75: 66,
+    76: 67,
+    77: 68,
+    78: 69,
+    79: 70,
+    80: 71,
+    81: 72,
+    82: 73,
+    84: 74,
+    85: 75,
+    86: 76,
+    87: 77,
+    88: 78,
+    89: 79,
+    90: 80,
+    92: 81,
+    93: 82,
+    94: 83,
+    95: 84,
+    96: 85,
+    97: 86,
+    98: 87,
+    99: 88,
+    100: 89,
+    101: 90,
+    102: 91,
+    103: 92,
+    104: 93,
+    105: 94,
+    106: 95,
+    107: 96,
+    108: 97,
+    109: 98,
+    110: 99,
+    111: 100,
+    112: 101,
+    113: 102,
+    114: 103,
+    115: 104,
+    116: 105,
+    117: 106,
+    118: 107,
+    119: 108,
+    120: 109,
+    121: 110,
+    122: 111,
+    123: 112,
+    124: 113,
+    125: 114,
+    126: 115,
+    127: 116,
+    128: 117,
+    129: 118,
+    130: 119,
+    131: 120,
+    132: 121,
+    133: 122,
+    134: 123,
+    135: 124,
+    136: 125,
+    137: 126,
+    138: 127,
+    139: 128,
+    140: 129,
+    141: 130,
+    142: 131,
+    143: 132,
+    144: 133,
+    145: 134,
+    146: 135,
+    147: 136,
+    148: 137,
+    149: 138,
+    150: 139,
+    151: 140,
+    152: 141,
+    153: 142,
+    154: 143,
+    155: 144,
+    156: 145,
+    157: 146,
+    158: 147,
+    159: 148,
+    160: 149,
+    161: 150,
+    162: 151,
+    163: 152,
+    164: 153,
+    165: 154,
+    166: 155,
+    167: 156,
+    168: 157,
+    169: 158,
+    170: 159,
+    171: 160,
+    172: 161,
+    173: 162,
+    174: 163,
+    175: 164,
+    176: 165,
+    177: 166,
+    178: 167,
+    179: 168,
+    180: 169,
+    181: 170,
+    182: 171
+}
+
+
+def convert_to_trainID(tuple_path, in_img_dir, in_ann_dir, out_img_dir,
+                       out_mask_dir, is_train):
+    imgpath, maskpath = tuple_path
+    shutil.copyfile(
+        osp.join(in_img_dir, imgpath),
+        osp.join(out_img_dir, 'train2014', imgpath) if is_train else osp.join(
+            out_img_dir, 'test2014', imgpath))
+    annotate = loadmat(osp.join(in_ann_dir, maskpath))
+    mask = annotate['S'].astype(np.uint8)
+    mask_copy = mask.copy()
+    for clsID, trID in clsID_to_trID.items():
+        mask_copy[mask == clsID] = trID
+    seg_filename = osp.join(out_mask_dir, 'train2014',
+                            maskpath.split('.')[0] +
+                            '_labelTrainIds.png') if is_train else osp.join(
+                                out_mask_dir, 'test2014',
+                                maskpath.split('.')[0] + '_labelTrainIds.png')
+    Image.fromarray(mask_copy).save(seg_filename, 'PNG')
+
+
+def generate_coco_list(folder):
+    train_list = osp.join(folder, 'imageLists', 'train.txt')
+    test_list = osp.join(folder, 'imageLists', 'test.txt')
+    train_paths = []
+    test_paths = []
+
+    with open(train_list) as f:
+        for filename in f:
+            basename = filename.strip()
+            imgpath = basename + '.jpg'
+            maskpath = basename + '.mat'
+            train_paths.append((imgpath, maskpath))
+
+    with open(test_list) as f:
+        for filename in f:
+            basename = filename.strip()
+            imgpath = basename + '.jpg'
+            maskpath = basename + '.mat'
+            test_paths.append((imgpath, maskpath))
+
+    return train_paths, test_paths
+
+
+def parse_args():
+    parser = argparse.ArgumentParser(
+        description=\
+        'Convert COCO Stuff 10k annotations to mmsegmentation format')  # noqa
+    parser.add_argument('coco_path', help='coco stuff path')
+    parser.add_argument('-o', '--out_dir', help='output path')
+    parser.add_argument(
+        '--nproc', default=16, type=int, help='number of process')
+    args = parser.parse_args()
+    return args
+
+
+def main():
+    args = parse_args()
+    coco_path = args.coco_path
+    nproc = args.nproc
+
+    out_dir = args.out_dir or coco_path
+    out_img_dir = osp.join(out_dir, 'images')
+    out_mask_dir = osp.join(out_dir, 'annotations')
+
+    mmcv.mkdir_or_exist(osp.join(out_img_dir, 'train2014'))
+    mmcv.mkdir_or_exist(osp.join(out_img_dir, 'test2014'))
+    mmcv.mkdir_or_exist(osp.join(out_mask_dir, 'train2014'))
+    mmcv.mkdir_or_exist(osp.join(out_mask_dir, 'test2014'))
+
+    train_list, test_list = generate_coco_list(coco_path)
+    assert (len(train_list) +
+            len(test_list)) == COCO_LEN, 'Wrong length of list {} & {}'.format(
+                len(train_list), len(test_list))
+
+    if args.nproc > 1:
+        mmcv.track_parallel_progress(
+            partial(
+                convert_to_trainID,
+                in_img_dir=osp.join(coco_path, 'images'),
+                in_ann_dir=osp.join(coco_path, 'annotations'),
+                out_img_dir=out_img_dir,
+                out_mask_dir=out_mask_dir,
+                is_train=True),
+            train_list,
+            nproc=nproc)
+        mmcv.track_parallel_progress(
+            partial(
+                convert_to_trainID,
+                in_img_dir=osp.join(coco_path, 'images'),
+                in_ann_dir=osp.join(coco_path, 'annotations'),
+                out_img_dir=out_img_dir,
+                out_mask_dir=out_mask_dir,
+                is_train=False),
+            test_list,
+            nproc=nproc)
+    else:
+        mmcv.track_progress(
+            partial(
+                convert_to_trainID,
+                in_img_dir=osp.join(coco_path, 'images'),
+                in_ann_dir=osp.join(coco_path, 'annotations'),
+                out_img_dir=out_img_dir,
+                out_mask_dir=out_mask_dir,
+                is_train=True), train_list)
+        mmcv.track_progress(
+            partial(
+                convert_to_trainID,
+                in_img_dir=osp.join(coco_path, 'images'),
+                in_ann_dir=osp.join(coco_path, 'annotations'),
+                out_img_dir=out_img_dir,
+                out_mask_dir=out_mask_dir,
+                is_train=False), test_list)
+
+    print('Done!')
+
+
+if __name__ == '__main__':
+    main()

tools/convert_datasets/coco_stuff164k.py

@@ -0,0 +1,263 @@
+import argparse
+import os.path as osp
+import shutil
+from functools import partial
+from glob import glob
+
+import mmcv
+import numpy as np
+from PIL import Image
+
+COCO_LEN = 123287
+
+clsID_to_trID = {
+    0: 0,
+    1: 1,
+    2: 2,
+    3: 3,
+    4: 4,
+    5: 5,
+    6: 6,
+    7: 7,
+    8: 8,
+    9: 9,
+    10: 10,
+    12: 11,
+    13: 12,
+    14: 13,
+    15: 14,
+    16: 15,
+    17: 16,
+    18: 17,
+    19: 18,
+    20: 19,
+    21: 20,
+    22: 21,
+    23: 22,
+    24: 23,
+    26: 24,
+    27: 25,
+    30: 26,
+    31: 27,
+    32: 28,
+    33: 29,
+    34: 30,
+    35: 31,
+    36: 32,
+    37: 33,
+    38: 34,
+    39: 35,
+    40: 36,
+    41: 37,
+    42: 38,
+    43: 39,
+    45: 40,
+    46: 41,
+    47: 42,
+    48: 43,
+    49: 44,
+    50: 45,
+    51: 46,
+    52: 47,
+    53: 48,
+    54: 49,
+    55: 50,
+    56: 51,
+    57: 52,
+    58: 53,
+    59: 54,
+    60: 55,
+    61: 56,
+    62: 57,
+    63: 58,
+    64: 59,
+    66: 60,
+    69: 61,
+    71: 62,
+    72: 63,
+    73: 64,
+    74: 65,
+    75: 66,
+    76: 67,
+    77: 68,
+    78: 69,
+    79: 70,
+    80: 71,
+    81: 72,
+    83: 73,
+    84: 74,
+    85: 75,
+    86: 76,
+    87: 77,
+    88: 78,
+    89: 79,
+    91: 80,
+    92: 81,
+    93: 82,
+    94: 83,
+    95: 84,
+    96: 85,
+    97: 86,
+    98: 87,
+    99: 88,
+    100: 89,
+    101: 90,
+    102: 91,
+    103: 92,
+    104: 93,
+    105: 94,
+    106: 95,
+    107: 96,
+    108: 97,
+    109: 98,
+    110: 99,
+    111: 100,
+    112: 101,
+    113: 102,
+    114: 103,
+    115: 104,
+    116: 105,
+    117: 106,
+    118: 107,
+    119: 108,
+    120: 109,
+    121: 110,
+    122: 111,
+    123: 112,
+    124: 113,
+    125: 114,
+    126: 115,
+    127: 116,
+    128: 117,
+    129: 118,
+    130: 119,
+    131: 120,
+    132: 121,
+    133: 122,
+    134: 123,
+    135: 124,
+    136: 125,
+    137: 126,
+    138: 127,
+    139: 128,
+    140: 129,
+    141: 130,
+    142: 131,
+    143: 132,
+    144: 133,
+    145: 134,
+    146: 135,
+    147: 136,
+    148: 137,
+    149: 138,
+    150: 139,
+    151: 140,
+    152: 141,
+    153: 142,
+    154: 143,
+    155: 144,
+    156: 145,
+    157: 146,
+    158: 147,
+    159: 148,
+    160: 149,
+    161: 150,
+    162: 151,
+    163: 152,
+    164: 153,
+    165: 154,
+    166: 155,
+    167: 156,
+    168: 157,
+    169: 158,
+    170: 159,
+    171: 160,
+    172: 161,
+    173: 162,
+    174: 163,
+    175: 164,
+    176: 165,
+    177: 166,
+    178: 167,
+    179: 168,
+    180: 169,
+    181: 170,
+    255: 255
+}
+
+
+def convert_to_trainID(maskpath, out_mask_dir, is_train):
+    mask = np.array(Image.open(maskpath))
+    mask_copy = mask.copy()
+    for clsID, trID in clsID_to_trID.items():
+        mask_copy[mask == clsID] = trID
+    seg_filename = osp.join(
+        out_mask_dir, 'train2017',
+        osp.basename(maskpath).split('.')[0] +
+        '_labelTrainIds.png') if is_train else osp.join(
+            out_mask_dir, 'val2017',
+            osp.basename(maskpath).split('.')[0] + '_labelTrainIds.png')
+    Image.fromarray(mask_copy).save(seg_filename, 'PNG')
+
+
+def parse_args():
+    parser = argparse.ArgumentParser(
+        description=\
+        'Convert COCO Stuff 164k annotations to mmsegmentation format')  # noqa
+    parser.add_argument('coco_path', help='coco stuff path')
+    parser.add_argument('-o', '--out_dir', help='output path')
+    parser.add_argument(
+        '--nproc', default=16, type=int, help='number of process')
+    args = parser.parse_args()
+    return args
+
+
+def main():
+    args = parse_args()
+    coco_path = args.coco_path
+    nproc = args.nproc
+
+    out_dir = args.out_dir or coco_path
+    out_img_dir = osp.join(out_dir, 'images')
+    out_mask_dir = osp.join(out_dir, 'annotations')
+
+    mmcv.mkdir_or_exist(osp.join(out_mask_dir, 'train2017'))
+    mmcv.mkdir_or_exist(osp.join(out_mask_dir, 'val2017'))
+
+    if out_dir != coco_path:
+        shutil.copytree(osp.join(coco_path, 'images'), out_img_dir)
+
+    train_list = glob(osp.join(coco_path, 'annotations', 'train2017', '*.png'))
+    train_list = [file for file in train_list if '_labelTrainIds' not in file]
+    test_list = glob(osp.join(coco_path, 'annotations', 'val2017', '*.png'))
+    test_list = [file for file in test_list if '_labelTrainIds' not in file]
+    assert (len(train_list) +
+            len(test_list)) == COCO_LEN, 'Wrong length of list {} & {}'.format(
+                len(train_list), len(test_list))
+
+    if args.nproc > 1:
+        mmcv.track_parallel_progress(
+            partial(
+                convert_to_trainID, out_mask_dir=out_mask_dir, is_train=True),
+            train_list,
+            nproc=nproc)
+        mmcv.track_parallel_progress(
+            partial(
+                convert_to_trainID, out_mask_dir=out_mask_dir, is_train=False),
+            test_list,
+            nproc=nproc)
+    else:
+        mmcv.track_progress(
+            partial(
+                convert_to_trainID, out_mask_dir=out_mask_dir, is_train=True),
+            train_list)
+        mmcv.track_progress(
+            partial(
+                convert_to_trainID, out_mask_dir=out_mask_dir, is_train=False),
+            test_list)
+
+    print('Done!')
+
+
+if __name__ == '__main__':
+    main()