def ArraysToDelfFeatures(locations,
scales,
descriptors,
attention,
orientations=None):
num_features = len(attention)
assert num_features == locations.shape[0]
assert num_features == len(scales)
assert num_features == descriptors.shape[0]
if orientations is None:
orientations = np.zeros([num_features], dtype=np.float32)
else:
assert num_features == len(orientations)
delf_features = feature_pb2.DelfFeatures()
for i in xrange(num_features):
delf_feature = delf_features.feature.add()
delf_feature.y = locations[i, 0]
delf_feature.x = locations[i, 1]
delf_feature.scale = scales[i]
delf_feature.orientation = orientations[i]
delf_feature.strength = attention[i]
delf_feature.descriptor.CopyFrom(
datum_io.ArrayToDatum(descriptors[i, ]))
return delf_features
def ArraysToDelfFeatures(locations,
scales,
descriptors,
attention,
orientations=None):
"""Converts DELF features to DelfFeatures proto.
Args:
locations: [N, 2] float array which denotes the selected keypoint
locations. N is the number of features.
scales: [N] float array with feature scales.
descriptors: [N, depth] float array with DELF descriptors.
attention: [N] float array with attention scores.
orientations: [N] float array with orientations. If None, all orientations
are set to zero.
Returns:
delf_features: DelfFeatures object.
"""
num_features = len(attention)
assert num_features == locations.shape[0]
assert num_features == len(scales)
assert num_features == descriptors.shape[0]
if orientations is None:
orientations = np.zeros([num_features], dtype=np.float32)
else:
assert num_features == len(orientations)
delf_features = feature_pb2.DelfFeatures()
for i in range(num_features):
delf_feature = delf_features.feature.add()
delf_feature.y = locations[i, 0]
delf_feature.x = locations[i, 1]
delf_feature.scale = scales[i]
delf_feature.orientation = orientations[i]
delf_feature.strength = attention[i]
delf_feature.descriptor.CopyFrom(
datum_io.ArrayToDatum(descriptors[i, ]))
return delf_features