def __init__(self, image_set, devkit_path, data_path):
imdb.__init__(self, image_set)
self._image_set = image_set
self._devkit_path = devkit_path
self._data_path = data_path
synsets_image = sio.loadmat(
os.path.join(self._devkit_path, 'data', 'meta_det.mat'))
synsets_video = sio.loadmat(
os.path.join(self._devkit_path, 'data', 'meta_vid.mat'))
self._classes_image = ('__background__', )
self._wnid_image = (0, )
self._classes = ('__background__', )
self._wnid = (0, )
for i in xrange(200):
self._classes_image = self._classes_image + (
synsets_image['synsets'][0][i][2][0], )
self._wnid_image = self._wnid_image + (
synsets_image['synsets'][0][i][1][0], )
for i in xrange(30):
self._classes = self._classes + (
synsets_video['synsets'][0][i][2][0], )
self._wnid = self._wnid + (synsets_video['synsets'][0][i][1][0], )
self._wnid_to_ind_image = dict(zip(self._wnid_image, xrange(201)))
self._class_to_ind_image = dict(zip(self._classes_image, xrange(201)))
self._wnid_to_ind = dict(zip(self._wnid, xrange(31)))
self._class_to_ind = dict(zip(self._classes, xrange(31)))
#check for valid intersection between video and image classes
self._valid_image_flag = [0] * 201
for i in range(1, 201):
if self._wnid_image[i] in self._wnid_to_ind:
self._valid_image_flag[i] = 1
self._image_ext = ['.JPEG']
self._image_index = self._load_image_set_index()
# Default to roidb handler
self._roidb_handler = self.gt_roidb
# Specific config options
self.config = {'cleanup': True, 'use_salt': True, 'top_k': 2000}
assert os.path.exists(
self._devkit_path), 'Devkit path does not exist: {}'.format(
self._devkit_path)
assert os.path.exists(
self._data_path), 'Path does not exist: {}'.format(self._data_path)
def __init__(self, image_set, year, devkit_path=None):
imdb.__init__(self, 'voc_' + year + '_' + image_set)
self._year = year
self._image_set = image_set
self._devkit_path = self._get_default_path() if devkit_path is None \
else devkit_path
self._data_path = os.path.join(self._devkit_path, 'VOC' + self._year)
self._classes = (
'__background__', # always index 0
'aeroplane',
'bicycle',
'bird',
'boat',
'bottle',
'bus',
'car',
'cat',
'chair',
'cow',
'diningtable',
'dog',
'horse',
'motorbike',
'person',
'pottedplant',
'sheep',
'sofa',
'train',
'tvmonitor')
self._class_to_ind = dict(
list(zip(self.classes, list(range(self.num_classes)))))
self._image_ext = '.jpg'
self._image_index = self._load_image_set_index()
# Default to roidb handler
self._roidb_handler = self.gt_roidb
self._salt = str(uuid.uuid4())
self._comp_id = 'comp4'
# PASCAL specific config options
self.config = {
'cleanup': True,
'use_salt': True,
'use_diff': False,
'matlab_eval': False,
'rpn_file': None
}
assert os.path.exists(self._devkit_path), \
'VOCdevkit path does not exist: {}'.format(self._devkit_path)
assert os.path.exists(self._data_path), \
'Path does not exist: {}'.format(self._data_path)
def __init__(self, image_set, year):
imdb.__init__(self, 'coco_' + year + '_' + image_set)
# COCO specific config options
self.config = {'use_salt': True,
'cleanup': True}
# name, paths
self._year = year
self._image_set = image_set
self._data_path = osp.join(cfg.DATA_DIR, 'coco')
# load COCO API, classes, class <-> id mappings
self._COCO = COCO(self._get_ann_file())
cats = self._COCO.loadCats(self._COCO.getCatIds())
self._classes = tuple(['__background__'] + [c['name'] for c in cats])
self._class_to_ind = dict(list(zip(self.classes, list(range(self.num_classes)))))
self._class_to_coco_cat_id = dict(list(zip([c['name'] for c in cats],
self._COCO.getCatIds())))
self._image_index = self._load_image_set_index()
# Default to roidb handler
self.set_proposal_method('gt')
self.competition_mode(False)
# Some image sets are "views" (i.e. subsets) into others.
# For example, minival2014 is a random 5000 image subset of val2014.
# This mapping tells us where the view's images and proposals come from.
self._view_map = {
'minival2014': 'val2014', # 5k val2014 subset
'valminusminival2014': 'val2014', # val2014 \setminus minival2014
'test-dev2015': 'test2015',
'valminuscapval2014': 'val2014',
'capval2014': 'val2014',
'captest2014': 'val2014'
}
coco_name = image_set + year # e.g., "val2014"
self._data_name = (self._view_map[coco_name]
if coco_name in self._view_map
else coco_name)
# Dataset splits that have ground-truth annotations (test splits
# do not have gt annotations)
self._gt_splits = ('train', 'val', 'minival')
def __init__(self, version, image_set, xml_path):
imdb.__init__(self, 'vg_' + version + '_' + image_set)
self._version = version
self.xml_path = xml_path
self._image_set = image_set
self._data_path = os.path.join(cfg.DATA_DIR, 'genome')
self._img_path = os.path.join(cfg.DATA_DIR, 'vg')
# VG specific config options
self.config = {'cleanup' : False}
# Load classes
self._classes = ['__background__']
self._class_to_ind = {}
self._class_to_ind[self._classes[0]] = 0
with open(os.path.join(self._data_path, self._version, 'objects_vocab.txt')) as f:
count = 1
for object in f.readlines():
names = [n.lower().strip() for n in object.split(',')]
self._classes.append(names[0])
for n in names:
self._class_to_ind[n] = count
count += 1
# Load attributes
self._attributes = ['__no_attribute__']
self._attribute_to_ind = {}
self._attribute_to_ind[self._attributes[0]] = 0
with open(os.path.join(self._data_path, self._version, 'attributes_vocab.txt')) as f:
count = 1
for att in f.readlines():
names = [n.lower().strip() for n in att.split(',')]
self._attributes.append(names[0])
for n in names:
self._attribute_to_ind[n] = count
count += 1
# Load relations
self._relations = ['__no_relation__']
self._relation_to_ind = {}
self._relation_to_ind[self._relations[0]] = 0
with open(os.path.join(self._data_path, self._version, 'relations_vocab.txt')) as f:
count = 1
for rel in f.readlines():
names = [n.lower().strip() for n in rel.split(',')]
self._relations.append(names[0])
for n in names:
self._relation_to_ind[n] = count
count += 1
self._image_ext = '.jpg'
load_index_from_file = False
if os.path.exists(os.path.join(self._data_path, "vg_image_index_{}.p".format(self._image_set))):
with open(os.path.join(self._data_path, "vg_image_index_{}.p".format(self._image_set)), 'rb') as fp:
self._image_index = pickle.load(fp)
load_index_from_file = True
load_id_from_file = False
if os.path.exists(os.path.join(self._data_path, "vg_id_to_dir_{}.p".format(self._image_set))):
with open(os.path.join(self._data_path, "vg_id_to_dir_{}.p".format(self._image_set)), 'rb') as fp:
self._id_to_dir = pickle.load(fp)
load_id_from_file = True
if not load_index_from_file or not load_id_from_file:
self._image_index, self._id_to_dir = self._load_image_set_index()
with open(os.path.join(self._data_path, "vg_image_index_{}.p".format(self._image_set)), 'wb') as fp:
pickle.dump(self._image_index, fp)
with open(os.path.join(self._data_path, "vg_id_to_dir_{}.p".format(self._image_set)), 'wb') as fp:
pickle.dump(self._id_to_dir, fp)
self._roidb_handler = self.gt_roidb