Add support of the wider face dataset (#765)

* Add support of the wider face dataset

* Update readme

* Prevent ignoring boxes in test mode

* Rename wider to wider_face

* Rename configs for wider face

configs/wider_face/README.md

+## WIDER Face Dataset
+
+To use the WIDER Face dataset you need to download it
+and extract to the `data/WIDERFace` folder. Annotation in the VOC format
+can be found in this [repo](https://github.com/sovrasov/wider-face-pascal-voc-annotations.git).
+You should move the annotation files from `WIDER_train_annotations` and `WIDER_val_annotations` folders
+to the `Annotation` folders inside the corresponding directories `WIDER_train` and `WIDER_val`.
+Also annotation lists `val.txt` and `train.txt` should be copied to `data/WIDERFace` from `WIDER_train_annotations` and `WIDER_val_annotations`.
+The directory should be like this:
+
+```
+mmdetection
+├── mmdet
+├── tools
+├── configs
+├── data
+│   ├── WIDERFace
+│   │   ├── WIDER_train
+│   |   │   ├──0--Parade
+│   |   │   ├── ...
+│   |   │   ├── Annotations
+│   │   ├── WIDER_val
+│   |   │   ├──0--Parade
+│   |   │   ├── ...
+│   |   │   ├── Annotations
+│   │   ├── val.txt
+│   │   ├── train.txt
+
+```
+
+After that you can train the SSD300 on WIDER by launching training with the `ssd300_wider_face.py` config or
+create your own config based on the presented one.

configs/wider_face/ssd300_wider_face.py

+# model settings
+input_size = 300
+model = dict(
+    type='SingleStageDetector',
+    pretrained='open-mmlab://vgg16_caffe',
+    backbone=dict(
+        type='SSDVGG',
+        input_size=input_size,
+        depth=16,
+        with_last_pool=False,
+        ceil_mode=True,
+        out_indices=(3, 4),
+        out_feature_indices=(22, 34),
+        l2_norm_scale=20),
+    neck=None,
+    bbox_head=dict(
+        type='SSDHead',
+        input_size=input_size,
+        in_channels=(512, 1024, 512, 256, 256, 256),
+        num_classes=2,
+        anchor_strides=(8, 16, 32, 64, 100, 300),
+        basesize_ratio_range=(0.15, 0.9),
+        anchor_ratios=([2], [2, 3], [2, 3], [2, 3], [2], [2]),
+        target_means=(.0, .0, .0, .0),
+        target_stds=(0.1, 0.1, 0.2, 0.2)))
+cudnn_benchmark = True
+train_cfg = dict(
+    assigner=dict(
+        type='MaxIoUAssigner',
+        pos_iou_thr=0.5,
+        neg_iou_thr=0.5,
+        min_pos_iou=0.,
+        ignore_iof_thr=-1,
+        gt_max_assign_all=False),
+    smoothl1_beta=1.,
+    allowed_border=-1,
+    pos_weight=-1,
+    neg_pos_ratio=3,
+    debug=False)
+test_cfg = dict(
+    nms=dict(type='nms', iou_thr=0.45),
+    min_bbox_size=0,
+    score_thr=0.02,
+    max_per_img=200)
+# model training and testing settings
+# dataset settings
+dataset_type = 'WIDERFaceDataset'
+data_root = 'data/WIDERFace/'
+img_norm_cfg = dict(mean=[123.675, 116.28, 103.53], std=[1, 1, 1], to_rgb=True)
+data = dict(
+    imgs_per_gpu=60,
+    workers_per_gpu=2,
+    train=dict(
+        type='RepeatDataset',
+        times=2,
+        dataset=dict(
+            type=dataset_type,
+            ann_file=[
+                data_root + 'train.txt',
+            ],
+            img_prefix=[data_root + 'WIDER_train/'],
+            img_scale=(300, 300),
+            min_size=17,  # throw away very small faces to improve training,
+            # because 300x300 is too low resolution to detect them
+            img_norm_cfg=img_norm_cfg,
+            size_divisor=None,
+            flip_ratio=0.5,
+            with_mask=False,
+            with_crowd=False,
+            with_label=True,
+            test_mode=False,
+            extra_aug=dict(
+                photo_metric_distortion=dict(
+                    brightness_delta=32,
+                    contrast_range=(0.5, 1.5),
+                    saturation_range=(0.5, 1.5),
+                    hue_delta=18),
+                expand=dict(
+                    mean=img_norm_cfg['mean'],
+                    to_rgb=img_norm_cfg['to_rgb'],
+                    ratio_range=(1, 4)),
+                random_crop=dict(
+                    min_ious=(0.1, 0.3, 0.5, 0.7, 0.9), min_crop_size=0.3)),
+            resize_keep_ratio=False)),
+    val=dict(
+        type=dataset_type,
+        ann_file=data_root + '/val.txt',
+        img_prefix=data_root + 'WIDER_val/',
+        img_scale=(300, 300),
+        img_norm_cfg=img_norm_cfg,
+        size_divisor=None,
+        flip_ratio=0,
+        with_mask=False,
+        with_label=False,
+        test_mode=True,
+        resize_keep_ratio=False),
+    test=dict(
+        type=dataset_type,
+        ann_file=data_root + '/val.txt',
+        img_prefix=data_root + 'WIDER_val/',
+        img_scale=(300, 300),
+        img_norm_cfg=img_norm_cfg,
+        size_divisor=None,
+        flip_ratio=0,
+        with_mask=False,
+        with_label=False,
+        test_mode=True,
+        resize_keep_ratio=False))
+# optimizer
+optimizer = dict(type='SGD', lr=1e-3, momentum=0.9, weight_decay=5e-4)
+optimizer_config = dict()
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=1000,
+    warmup_ratio=1.0 / 3,
+    step=[16, 20])
+checkpoint_config = dict(interval=1)
+# yapf:disable
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        dict(type='TensorboardLoggerHook')
+    ])
+# yapf:enable
+# runtime settings
+total_epochs = 24
+dist_params = dict(backend='nccl')
+log_level = 'INFO'
+work_dir = './work_dirs/ssd300_wider'
+load_from = None
+resume_from = None
+workflow = [('train', 1)]

mmdet/core/evaluation/class_names.py

 import mmcv
 
 
+def wider_face_classes():
+    return ['face']
+
+
 def voc_classes():
     return [
         'aeroplane', 'bicycle', 'bird', 'boat', 'bottle', 'bus', 'car', 'cat',
@@ -82,7 +86,8 @@ dataset_aliases = {
     'voc': ['voc', 'pascal_voc', 'voc07', 'voc12'],
     'imagenet_det': ['det', 'imagenet_det', 'ilsvrc_det'],
     'imagenet_vid': ['vid', 'imagenet_vid', 'ilsvrc_vid'],
-    'coco': ['coco', 'mscoco', 'ms_coco']
+    'coco': ['coco', 'mscoco', 'ms_coco'],
+    'wider_face': ['WIDERFaceDataset', 'wider_face', 'WDIERFace']
 }
 
 

mmdet/datasets/__init__.py

@@ -2,6 +2,7 @@ from .custom import CustomDataset
 from .xml_style import XMLDataset
 from .coco import CocoDataset
 from .voc import VOCDataset
+from .wider_face import WIDERFaceDataset
 from .loader import GroupSampler, DistributedGroupSampler, build_dataloader
 from .utils import to_tensor, random_scale, show_ann, get_dataset
 from .concat_dataset import ConcatDataset
@@ -12,5 +13,5 @@ __all__ = [
     'CustomDataset', 'XMLDataset', 'CocoDataset', 'VOCDataset', 'GroupSampler',
     'DistributedGroupSampler', 'build_dataloader', 'to_tensor', 'random_scale',
     'show_ann', 'get_dataset', 'ConcatDataset', 'RepeatDataset',
-    'ExtraAugmentation'
+    'ExtraAugmentation', 'WIDERFaceDataset'
 ]

mmdet/datasets/wider_face.py

+import os.path as osp
+import xml.etree.ElementTree as ET
+
+import mmcv
+
+from .xml_style import XMLDataset
+
+
+class WIDERFaceDataset(XMLDataset):
+    """
+    Reader for the WIDER Face dataset in PASCAL VOC format.
+    Conversion scripts can be found in
+    https://github.com/sovrasov/wider-face-pascal-voc-annotations
+    """
+    CLASSES = ('face',)
+
+    def __init__(self, **kwargs):
+        super(WIDERFaceDataset, self).__init__(**kwargs)
+
+    def load_annotations(self, ann_file):
+        img_infos = []
+        img_ids = mmcv.list_from_file(ann_file)
+        for img_id in img_ids:
+            filename = '{}.jpg'.format(img_id)
+            xml_path = osp.join(self.img_prefix, 'Annotations',
+                                '{}.xml'.format(img_id))
+            tree = ET.parse(xml_path)
+            root = tree.getroot()
+            size = root.find('size')
+            width = int(size.find('width').text)
+            height = int(size.find('height').text)
+            folder = root.find('folder').text
+            img_infos.append(
+                dict(id=img_id, filename=osp.join(folder, filename),
+                     width=width, height=height))
+
+        return img_infos

mmdet/datasets/xml_style.py

@@ -9,9 +9,10 @@ from .custom import CustomDataset
 
 class XMLDataset(CustomDataset):
 
-    def __init__(self, **kwargs):
+    def __init__(self, min_size=None, **kwargs):
         super(XMLDataset, self).__init__(**kwargs)
         self.cat2label = {cat: i + 1 for i, cat in enumerate(self.CLASSES)}
+        self.min_size = min_size
 
     def load_annotations(self, ann_file):
         img_infos = []
@@ -50,7 +51,14 @@ class XMLDataset(CustomDataset):
                 int(bnd_box.find('xmax').text),
                 int(bnd_box.find('ymax').text)
             ]
-            if difficult:
+            ignore = False
+            if self.min_size:
+                assert not self.test_mode
+                w = bbox[2] - bbox[0]
+                h = bbox[3] - bbox[1]
+                if w < self.min_size or h < self.min_size:
+                    ignore = True
+            if difficult or ignore:
                 bboxes_ignore.append(bbox)
                 labels_ignore.append(label)
             else: