[Project] Support CAT-Seg from CVPR2023

configs/_base_/datasets/ade20k_384x384.py

@@ -0,0 +1,68 @@
+# dataset settings
+dataset_type = 'ADE20KDataset'
+data_root = 'data/ade/ADEChallengeData2016'
+crop_size = (384, 384)
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', reduce_zero_label=True),
+    dict(
+        type='RandomResize',
+        scale=(2048, 512),
+        ratio_range=(0.5, 2.0),
+        keep_ratio=True),
+    dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75),
+    dict(type='RandomFlip', prob=0.5),
+    dict(type='PhotoMetricDistortion'),
+    dict(type='PackSegInputs')
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='Resize', scale=(2048, 512), keep_ratio=True),
+    # add loading annotation after ``Resize`` because ground truth
+    # does not need to do resize data transform
+    dict(type='LoadAnnotations', reduce_zero_label=True),
+    dict(type='PackSegInputs')
+]
+img_ratios = [0.5, 0.75, 1.0, 1.25, 1.5, 1.75]
+tta_pipeline = [
+    dict(type='LoadImageFromFile', backend_args=None),
+    dict(
+        type='TestTimeAug',
+        transforms=[
+            [
+                dict(type='Resize', scale_factor=r, keep_ratio=True)
+                for r in img_ratios
+            ],
+            [
+                dict(type='RandomFlip', prob=0., direction='horizontal'),
+                dict(type='RandomFlip', prob=1., direction='horizontal')
+            ], [dict(type='LoadAnnotations')], [dict(type='PackSegInputs')]
+        ])
+]
+train_dataloader = dict(
+    batch_size=4,
+    num_workers=4,
+    persistent_workers=True,
+    sampler=dict(type='InfiniteSampler', shuffle=True),
+    dataset=dict(
+        type=dataset_type,
+        data_root=data_root,
+        data_prefix=dict(
+            img_path='images/training', seg_map_path='annotations/training'),
+        pipeline=train_pipeline))
+val_dataloader = dict(
+    batch_size=1,
+    num_workers=4,
+    persistent_workers=True,
+    sampler=dict(type='DefaultSampler', shuffle=False),
+    dataset=dict(
+        type=dataset_type,
+        data_root=data_root,
+        data_prefix=dict(
+            img_path='images/validation',
+            seg_map_path='annotations/validation'),
+        pipeline=test_pipeline))
+test_dataloader = val_dataloader
+
+val_evaluator = dict(type='IoUMetric', iou_metrics=['mIoU'])
+test_evaluator = val_evaluator

configs/_base_/datasets/coco-stuff164k.py

@@ -48,7 +48,7 @@
         type=dataset_type,
         data_root=data_root,
         data_prefix=dict(
-            img_path='images/train2017', seg_map_path='annotations/val2017'),
+            img_path='images/train2017', seg_map_path='annotations/train2017'),
         pipeline=train_pipeline))
 val_dataloader = dict(
     batch_size=1,

configs/_base_/datasets/coco-stuff164k_384x384.py

@@ -0,0 +1,62 @@
+# dataset settings
+dataset_type = 'COCOStuffDataset'
+data_root = 'data/coco_stuff164k'
+crop_size = (384, 384)
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations'),
+    dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75),
+    dict(type='RandomFlip', prob=0.5),
+    dict(type='PhotoMetricDistortion'),
+    dict(type='PackSegInputs')
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='Resize', scale=(2048, 512), keep_ratio=True),
+    # add loading annotation after ``Resize`` because ground truth
+    # does not need to do resize data transform
+    dict(type='LoadAnnotations'),
+    dict(type='PackSegInputs')
+]
+img_ratios = [0.5, 0.75, 1.0, 1.25, 1.5, 1.75]
+tta_pipeline = [
+    dict(type='LoadImageFromFile', backend_args=None),
+    dict(
+        type='TestTimeAug',
+        transforms=[
+            [
+                dict(type='Resize', scale_factor=r, keep_ratio=True)
+                for r in img_ratios
+            ],
+            [
+                dict(type='RandomFlip', prob=0., direction='horizontal'),
+                dict(type='RandomFlip', prob=1., direction='horizontal')
+            ], [dict(type='LoadAnnotations')], [dict(type='PackSegInputs')]
+        ])
+]
+train_dataloader = dict(
+    batch_size=2,
+    num_workers=4,
+    persistent_workers=True,
+    sampler=dict(type='InfiniteSampler', shuffle=True),
+    dataset=dict(
+        type=dataset_type,
+        data_root=data_root,
+        data_prefix=dict(
+            img_path='images/train2017', seg_map_path='annotations/train2017'),
+        pipeline=train_pipeline))
+val_dataloader = dict(
+    batch_size=1,
+    num_workers=4,
+    persistent_workers=True,
+    sampler=dict(type='DefaultSampler', shuffle=False),
+    dataset=dict(
+        type=dataset_type,
+        data_root=data_root,
+        data_prefix=dict(
+            img_path='images/val2017', seg_map_path='annotations/val2017'),
+        pipeline=test_pipeline))
+test_dataloader = val_dataloader
+
+val_evaluator = dict(type='IoUMetric', iou_metrics=['mIoU'])
+test_evaluator = val_evaluator

configs/_base_/datasets/pascal_context_59_384x384.py

@@ -0,0 +1,72 @@
+# dataset settings
+dataset_type = 'PascalContextDataset59'
+data_root = 'data/VOCdevkit/VOC2010/'
+
+img_scale = (520, 520)
+crop_size = (384, 384)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', reduce_zero_label=True),
+    dict(
+        type='RandomResize',
+        scale=img_scale,
+        ratio_range=(0.5, 2.0),
+        keep_ratio=True),
+    dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75),
+    dict(type='RandomFlip', prob=0.5),
+    dict(type='PhotoMetricDistortion'),
+    dict(type='PackSegInputs')
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='Resize', scale=img_scale, keep_ratio=True),
+    # add loading annotation after ``Resize`` because ground truth
+    # does not need to do resize data transform
+    dict(type='LoadAnnotations', reduce_zero_label=True),
+    dict(type='PackSegInputs')
+]
+img_ratios = [0.5, 0.75, 1.0, 1.25, 1.5, 1.75]
+tta_pipeline = [
+    dict(type='LoadImageFromFile', backend_args=None),
+    dict(
+        type='TestTimeAug',
+        transforms=[
+            [
+                dict(type='Resize', scale_factor=r, keep_ratio=True)
+                for r in img_ratios
+            ],
+            [
+                dict(type='RandomFlip', prob=0., direction='horizontal'),
+                dict(type='RandomFlip', prob=1., direction='horizontal')
+            ], [dict(type='LoadAnnotations')], [dict(type='PackSegInputs')]
+        ])
+]
+train_dataloader = dict(
+    batch_size=4,
+    num_workers=4,
+    persistent_workers=True,
+    sampler=dict(type='InfiniteSampler', shuffle=True),
+    dataset=dict(
+        type=dataset_type,
+        data_root=data_root,
+        data_prefix=dict(
+            img_path='JPEGImages', seg_map_path='SegmentationClassContext'),
+        ann_file='ImageSets/SegmentationContext/train.txt',
+        pipeline=train_pipeline))
+val_dataloader = dict(
+    batch_size=1,
+    num_workers=4,
+    persistent_workers=True,
+    sampler=dict(type='DefaultSampler', shuffle=False),
+    dataset=dict(
+        type=dataset_type,
+        data_root=data_root,
+        data_prefix=dict(
+            img_path='JPEGImages', seg_map_path='SegmentationClassContext'),
+        ann_file='ImageSets/SegmentationContext/val.txt',
+        pipeline=test_pipeline))
+test_dataloader = val_dataloader
+
+val_evaluator = dict(type='IoUMetric', iou_metrics=['mIoU'])
+test_evaluator = val_evaluator