add fce ic15 (open-mmlab#258)

cuhk-hbsun · Jun 3, 2021 · 18e7ecc · 18e7ecc
1 parent 4882c8a
commit 18e7ecc
Show file tree

Hide file tree

Showing 6 changed files with 163 additions and 8 deletions.
diff --git a/configs/textdet/fcenet/README.md b/configs/textdet/fcenet/README.md
@@ -17,6 +17,12 @@
 
 ### CTW1500
 
-|                                     Method                             | Pretrained Model | Training set  |   Test set   | #epochs | Test size | Recall | Precision | Hmean |                                                                                        Download                                                                                                    |
-| :--------------------------------------------------------------------: | :--------------: | :-----------: | :----------: | :-----: | :-------: | :----: | :-------: | :---: | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
-| [FCENet](/configs/textdet/fcenet/fcenet_r50dcnv2_fpn_1500e_ctw1500.py) |     ImageNet     | CTW1500 Train | CTW1500 Test |  1500   |(736, 1080)| 0.828  |   0.875   | 0.851 | [model](https://download.openmmlab.com/mmocr/textdet/fcenet/fcenet_r50dcnv2_fpn_1500e_ctw1500-05d740bb.pth) \| [log](https://download.openmmlab.com/mmocr/textdet/fcenet/20210511_181328.log.json) |
+|                                     Method                             |     Backbone     | Pretrained Model | Training set  |   Test set   | #epochs | Test size | Recall | Precision | Hmean |                                                                                        Download                                                                                                    |
+| :--------------------------------------------------------------------: |:----------------:| :--------------: | :-----------: | :----------: | :-----: | :-------: | :----: | :-------: | :---: | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
+| [FCENet](/configs/textdet/fcenet/fcenet_r50dcnv2_fpn_1500e_ctw1500.py) | ResNet50 + DCNv2 |     ImageNet     | CTW1500 Train | CTW1500 Test |  1500   |(736, 1080)| 0.828  |   0.875   | 0.851 | [model](https://download.openmmlab.com/mmocr/textdet/fcenet/fcenet_r50dcnv2_fpn_1500e_ctw1500-05d740bb.pth) \| [log](https://download.openmmlab.com/mmocr/textdet/fcenet/20210511_181328.log.json) |
+
+### ICDAR2015
+
+|                                     Method                             |     Backbone     | Pretrained Model | Training set  |   Test set   | #epochs | Test size  | Recall | Precision | Hmean |                                                                                        Download                                                                                                    |
+| :--------------------------------------------------------------------: | :--------------: | :--------------: | :-----------: | :----------: | :-----: | :-------:  | :----: | :-------: | :---: | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
+| [FCENet](/configs/textdet/fcenet/fcenet_r50_fpn_1500e_icdar2015.py)    |     ResNet50     |     ImageNet     |   IC15 Train  |   IC15 Test  |  1500   |(2260, 2260)| 0.819  |   0.880   | 0.849 | [model](https://download.openmmlab.com/mmocr/textdet/fcenet/fcenet_r50dcnv2_fpn_1500e_ctw1500-05d740bb.pth) \| [log](https://download.openmmlab.com/mmocr/textdet/fcenet/20210511_181328.log.json) |
diff --git a/configs/textdet/fcenet/fcenet_r50_fpn_1500e_icdar2015.py b/configs/textdet/fcenet/fcenet_r50_fpn_1500e_icdar2015.py
@@ -0,0 +1,135 @@
+fourier_degree = 5
+model = dict(
+    type='FCENet',
+    pretrained='torchvision://resnet50',
+    backbone=dict(
+        type='ResNet',
+        depth=50,
+        num_stages=4,
+        out_indices=(1, 2, 3),
+        frozen_stages=-1,
+        norm_cfg=dict(type='BN', requires_grad=True),
+        norm_eval=False,
+        style='pytorch'),
+    neck=dict(
+        type='FPN',
+        in_channels=[512, 1024, 2048],
+        out_channels=256,
+        add_extra_convs=True,
+        extra_convs_on_inputs=False,  # use P5
+        num_outs=3,
+        relu_before_extra_convs=True,
+        act_cfg=None),
+    bbox_head=dict(
+        type='FCEHead',
+        in_channels=256,
+        scales=(8, 16, 32),
+        loss=dict(type='FCELoss'),
+        alpha=1.2,
+        beta=1.0,
+        text_repr_type='quad',
+        fourier_degree=fourier_degree,
+    ))
+
+train_cfg = None
+test_cfg = None
+
+dataset_type = 'IcdarDataset'
+data_root = 'data/icdar2015/'
+
+img_norm_cfg = dict(
+    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='LoadTextAnnotations',
+        with_bbox=True,
+        with_mask=True,
+        poly2mask=False),
+    dict(
+        type='ColorJitter',
+        brightness=32.0 / 255,
+        saturation=0.5,
+        contrast=0.5),
+    dict(type='Normalize', **img_norm_cfg),
+    dict(type='RandomScaling', size=800, scale=(3. / 4, 5. / 2)),
+    dict(
+        type='RandomCropFlip', crop_ratio=0.5, iter_num=1, min_area_ratio=0.2),
+    dict(
+        type='RandomCropPolyInstances',
+        instance_key='gt_masks',
+        crop_ratio=0.8,
+        min_side_ratio=0.3),
+    dict(
+        type='RandomRotatePolyInstances',
+        rotate_ratio=0.5,
+        max_angle=30,
+        pad_with_fixed_color=False),
+    dict(type='SquareResizePad', target_size=800, pad_ratio=0.6),
+    dict(type='RandomFlip', flip_ratio=0.5, direction='horizontal'),
+    dict(type='Pad', size_divisor=32),
+    dict(
+        type='FCENetTargets',
+        fourier_degree=fourier_degree,
+        level_proportion_range=((0, 0.4), (0.3, 0.7), (0.6, 1.0))),
+    dict(
+        type='CustomFormatBundle',
+        keys=['p3_maps', 'p4_maps', 'p5_maps'],
+        visualize=dict(flag=False, boundary_key=None)),
+    dict(type='Collect', keys=['img', 'p3_maps', 'p4_maps', 'p5_maps'])
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='MultiScaleFlipAug',
+        img_scale=(2260, 2260),
+        flip=False,
+        transforms=[
+            dict(type='Resize', img_scale=(1280, 800), keep_ratio=True),
+            dict(type='Normalize', **img_norm_cfg),
+            dict(type='Pad', size_divisor=32),
+            dict(type='ImageToTensor', keys=['img']),
+            dict(type='Collect', keys=['img']),
+        ])
+]
+data = dict(
+    samples_per_gpu=8,
+    workers_per_gpu=2,
+    train=dict(
+        type=dataset_type,
+        ann_file=data_root + '/instances_training.json',
+        img_prefix=data_root + '/imgs',
+        pipeline=train_pipeline),
+    val=dict(
+        type=dataset_type,
+        ann_file=data_root + '/instances_test.json',
+        img_prefix=data_root + '/imgs',
+        pipeline=test_pipeline),
+    test=dict(
+        type=dataset_type,
+        ann_file=data_root + '/instances_test.json',
+        img_prefix=data_root + '/imgs',
+        pipeline=test_pipeline))
+evaluation = dict(interval=5, metric='hmean-iou')
+
+# optimizer
+optimizer = dict(type='SGD', lr=1e-3, momentum=0.90, weight_decay=5e-4)
+optimizer_config = dict(grad_clip=None)
+lr_config = dict(policy='poly', power=0.9, min_lr=1e-7, by_epoch=True)
+total_epochs = 1500
+
+checkpoint_config = dict(interval=5)
+# yapf:disable
+log_config = dict(
+    interval=20,
+    hooks=[
+        dict(type='TextLoggerHook')
+
+    ])
+# yapf:enable
+dist_params = dict(backend='nccl')
+log_level = 'INFO'
+load_from = None
+resume_from = None
+workflow = [('train', 1)]
diff --git a/mmocr/datasets/pipelines/textdet_targets/fcenet_targets.py b/mmocr/datasets/pipelines/textdet_targets/fcenet_targets.py
@@ -300,7 +300,7 @@ def generate_level_targets(self, img_size, text_polys, ignore_polys):
 
             for ind, proportion_range in enumerate(lv_proportion_range):
                 if proportion_range[0] < proportion < proportion_range[1]:
-                    lv_text_polys[ind].append(
+                    lv_ignore_polys[ind].append(
                         [ignore_poly[0] / lv_size_divs[ind]])
 
         for ind, size_divisor in enumerate(lv_size_divs):

diff --git a/mmocr/models/textdet/dense_heads/fce_head.py b/mmocr/models/textdet/dense_heads/fce_head.py
@@ -43,6 +43,7 @@ def __init__(self,
                  nms_thr=0.1,
                  alpha=1.0,
                  beta=1.0,
+                 text_repr_type='poly',
                  train_cfg=None,
                  test_cfg=None):
 
@@ -63,6 +64,7 @@ def __init__(self,
         self.nms_thr = nms_thr
         self.alpha = alpha
         self.beta = beta
+        self.text_repr_type = text_repr_type
         self.train_cfg = train_cfg
         self.test_cfg = test_cfg
         self.out_channels_cls = 4
@@ -129,6 +131,6 @@ def _get_boundary_single(self, score_map, scale):
             scale=scale,
             alpha=self.alpha,
             beta=self.beta,
-            text_repr_type='poly',
+            text_repr_type=self.text_repr_type,
             score_thr=self.score_thr,
             nms_thr=self.nms_thr)
diff --git a/mmocr/models/textdet/postprocess/wrapper.py b/mmocr/models/textdet/postprocess/wrapper.py
@@ -419,7 +419,7 @@ def fcenet_decode(preds,
     """
     assert isinstance(preds, list)
     assert len(preds) == 2
-    assert text_repr_type == 'poly'
+    assert text_repr_type in ['poly', 'quad']
 
     cls_pred = preds[0][0]
     tr_pred = cls_pred[0:2].softmax(dim=0).data.cpu().numpy()
@@ -460,6 +460,16 @@ def fcenet_decode(preds,
         boundaries = boundaries + polygons
 
     boundaries = poly_nms(boundaries, nms_thr)
+
+    if text_repr_type == 'quad':
+        new_boundaries = []
+        for boundary in boundaries:
+            poly = np.array(boundary[:-1]).reshape(-1, 2).astype(np.float32)
+            score = boundary[-1]
+            points = cv2.boxPoints(cv2.minAreaRect(poly))
+            points = np.int0(points)
+            new_boundaries.append(points.reshape(-1).tolist() + [score])
+
     return boundaries
 
 

diff --git a/tests/test_models/test_detector.py b/tests/test_models/test_detector.py
@@ -372,8 +372,10 @@ def test_textsnake(cfg_file):
 
 
 @pytest.mark.skipif(not torch.cuda.is_available(), reason='requires cuda')
-@pytest.mark.parametrize(
-    'cfg_file', ['textdet/fcenet/fcenet_r50dcnv2_fpn_1500e_ctw1500.py'])
+@pytest.mark.parametrize('cfg_file', [
+    'textdet/fcenet/fcenet_r50dcnv2_fpn_1500e_ctw1500.py',
+    'textdet/fcenet/fcenet_r50_fpn_1500e_icdar2015.py'
+])
 def test_fcenet(cfg_file):
     model = _get_detector_cfg(cfg_file)
     model['pretrained'] = None