def test_iou_batched_gt_batched_anchor_normalized_coordinate():
# batch_size = 2, n_gt_boxes = 1
# [2, 1, 4]
gt_boxes = tf.constant([[[0.0, 0.5, 0.5, 1.0]], [[-0.25, 0.25, 0.25, 0.75]]])
# [2, 1, 4]
anchors = tf.constant([[[0.25, 0.25, 0.75, 0.75]], [[-0.5, 0.25, 0.0, 0.75]]])
iou_layer = IOUSimilarity()
similarity = iou_layer(gt_boxes, anchors)
expected_out = np.asarray([[1 / 7], [1 / 3]]).astype(np.float32).reshape((2, 1, 1))
np.testing.assert_allclose(expected_out, similarity)
def test_iou_batched_gt_batched_anchor_absolute_coordinate():
# batch_size = 2, n_gt_boxes = 1
# [2, 1, 4]
gt_boxes = tf.constant([[[0, 2, 2, 4]], [[-1, 1, 1, 3]]])
# [2, 1, 4]
anchors = tf.constant([[[1, 1, 3, 3]], [[-2, 1, 0, 3]]])
iou_layer = IOUSimilarity()
similarity = iou_layer(gt_boxes, anchors)
expected_out = np.asarray([[1 / 7], [1 / 3]]).astype(np.float32).reshape((2, 1, 1))
np.testing.assert_allclose(expected_out, similarity)
def test_iou_basic_normalized_coordinate():
# both gt box and two anchors are size 1
# the intersection between gt box and first anchor is 1 and union is 7
# the intersection between gt box and second anchor is 0 and union is 8
gt_boxes = tf.constant([[[0, 0.5, 0.5, 1.0]]])
anchors = tf.constant([[0.25, 0.25, 0.75, 0.75], [-0.5, 0.25, 0, 0.75]])
iou_layer = IOUSimilarity()
similarity = iou_layer(gt_boxes, anchors)
expected_out = np.asarray([1 / 7, 0]).astype(np.float32).reshape((1, 1, 2))
np.testing.assert_allclose(expected_out, similarity)
def test_iou_basic_absolute_coordinate():
# both gt box and two anchors are size 4
# the intersection between gt box and first anchor is 1 and union is 7
# the intersection between gt box and second anchor is 0 and union is 8
gt_boxes = tf.constant([[[0, 2, 2, 4]]])
anchors = tf.constant([[1, 1, 3, 3], [-2, 1, 0, 3]])
iou_layer = IOUSimilarity()
similarity = iou_layer(gt_boxes, anchors)
# batch_size = 1, n_gt_boxes = 1, n_anchors = 2
expected_out = np.asarray([1 / 7, 0]).astype(np.float32).reshape((1, 1, 2))
np.testing.assert_allclose(expected_out, similarity)
def test_ragged_gt_boxes_multi_anchor_absolute_coordinate():
# [2, ragged, 4]
gt_boxes = tf.ragged.constant(
[[[0, 2, 2, 4]], [[-1, 1, 1, 3], [-1, 1, 2, 3]]], ragged_rank=1
)
# [2, 4]
anchors = tf.constant([[1, 1, 3, 3], [-2, 1, 0, 3]])
iou_layer = IOUSimilarity()
similarity = iou_layer(gt_boxes, anchors)
expected_out = tf.ragged.constant([[[1 / 7, 0.0]], [[0.0, 1 / 3], [1 / 4, 1 / 4]]])
np.testing.assert_allclose(expected_out.values.numpy(), similarity.values.numpy())
def test_iou_multi_gt_multi_anchor_normalized_coordinate():
# batch_size = 1, n_gt_boxes = 2
# [1, 2, 4]
gt_boxes = tf.constant([[[0.0, 0.5, 0.5, 1.0], [-0.25, 0.25, 0.25, 0.75]]])
# [2, 4]
anchors = tf.constant([[0.25, 0.25, 0.75, 0.75], [-0.5, 0.25, 0.0, 0.75]])
iou_layer = IOUSimilarity()
similarity = iou_layer(gt_boxes, anchors)
expected_out = (
np.asarray([[1 / 7, 0], [0, 1 / 3]]).astype(np.float32).reshape((1, 2, 2))
)
np.testing.assert_allclose(expected_out, similarity)
def test_iou_multi_gt_multi_anchor_absolute_coordinate():
# batch_size = 1, n_gt_boxes = 2
# [1, 2, 4]
gt_boxes = tf.constant([[[0, 2, 2, 4], [-1, 1, 1, 3]]])
# [2, 4]
anchors = tf.constant([[1, 1, 3, 3], [-2, 1, 0, 3]])
iou_layer = IOUSimilarity()
similarity = iou_layer(gt_boxes, anchors)
expected_out = (
np.asarray([[1 / 7, 0], [0, 1 / 3]]).astype(np.float32).reshape((1, 2, 2))
)
np.testing.assert_allclose(expected_out, similarity)
def test_ragged_gt_boxes_empty_anchor():
# [2, ragged, 4]
gt_boxes = tf.ragged.constant(
[[[0.0, 0.5, 0.5, 1.0]], [[-0.25, 0.25, 0.25, 0.75], [-0.25, 0.25, 0.5, 0.75]]],
ragged_rank=1,
)
# [2, 4]
anchors = tf.constant([[0.25, 0.25, 0.25, 0.25], [-0.5, 0.25, 0.0, 0.75]])
iou_layer = IOUSimilarity()
similarity = iou_layer(gt_boxes, anchors)
expected_out = tf.ragged.constant([[[0.0, 0.0]], [[0.0, 1 / 3], [0.0, 1 / 4]]])
np.testing.assert_allclose(expected_out.values.numpy(), similarity.values.numpy())
def test_ragged_gt_boxes_batched_anchor_normalized_coordinate():
# [2, ragged, 4]
gt_boxes = tf.ragged.constant(
[[[0.0, 0.5, 0.5, 1.0]], [[-0.25, 0.25, 0.25, 0.75], [-0.25, 0.25, 0.5, 0.75]]],
ragged_rank=1,
)
# [2, 1, 4]
anchors = tf.constant([[[0.25, 0.25, 0.75, 0.75]], [[-0.5, 0.25, 0.0, 0.75]]])
iou_layer = IOUSimilarity()
similarity = iou_layer(gt_boxes, anchors)
expected_out = tf.ragged.constant([[[1 / 7]], [[1 / 3], [1 / 4]]])
np.testing.assert_allclose(expected_out.values.numpy(), similarity.values.numpy())
def test_iou_large():
# [2, 4]
gt_boxes = tf.constant([[4.0, 3.0, 7.0, 5.0], [5.0, 6.0, 10.0, 7.0]])
# [3, 4]
anchors = tf.constant(
[[3.0, 4.0, 6.0, 8.0], [14.0, 14.0, 15.0, 15.0], [0.0, 0.0, 20.0, 20.0]]
)
iou_layer = IOUSimilarity()
similarity = iou_layer(gt_boxes, anchors)
expected_out = np.asarray([[2 / 16, 0, 6 / 400], [1 / 16, 0.0, 5 / 400]]).astype(
np.float32
)
np.testing.assert_allclose(expected_out, similarity)
import tensorflow as tf
from kerascv.layers.iou_similarity import IOUSimilarity
iou_layer = IOUSimilarity()
class ArgMaxMatcher(tf.keras.layers.Layer):
"""ArgMax matcher"""
# [pos, neutral, neg]
def __init__(self, matched_threshold, unmatched_threshold):
self.matched_threshold = matched_threshold
self.unmatched_threshold = unmatched_threshold
super(ArgMaxMatcher, self).__init__()
# similarity: [#num_anchors, #num_gt_boxes]
# matched_values: [#num_gt_boxes, dim]
# unmatched_values: [dim]
# ignored_values: [dim]
def call(self, similarity, matched_values, unmatched_values, ignored_values):
# [#num_anchors]
matched_indices = tf.argmax(similarity, axis=1)
# [#num_anchors]
matched_max_vals = tf.reduce_max(similarity, axis=1)
above_unmatched_threshold_indices = tf.cast(
tf.greater(matched_max_vals, self.unmatched_threshold), tf.float32
)
# [#num_anchors]
below_unmatched_threshold_indices = tf.greater(
self.unmatched_threshold, matched_max_vals
# The anchors have areas of 32^2, 64^2, 128^2, 256^2, 521^2 on pyramid levels P3 to P7, respectively.
scales = [1, 2**(1 / 3), 2**(2 / 3)] * 3
aspect_ratios = [0.5, 1.0, 2.0] * 3
anchor_dimensions = [32, 64, 128, 256, 512]
anchor_strides = [2**3, 2**4, 2**5, 2**6, 2**7]
retina_resnet_scales = [scales] * 5
retina_resnet_aspect_ratios = [aspect_ratios] * 5
anchor_generator = MultiScaleAnchorGenerator(
scales=retina_resnet_scales,
aspect_ratios=retina_resnet_aspect_ratios,
dimensions=anchor_dimensions,
strides=anchor_strides)
anchors = anchor_generator([896, 896, 3])
similarity_cal = IOUSimilarity()
box_encoder = SSDBoxCoder(center_variances=[.1, .1], size_variances=[.2, .2])
box_decoder = SSDBoxCoder(center_variances=[.1, .1],
size_variances=[.2, .2],
invert=True)
retina_loss_layer = RetinaLossLayer(n_classes=80)
def resize_and_pad(image):
min_side = tf.cast(800, tf.float32)
max_side = tf.cast(1333, tf.float32)
stride = tf.cast(128.0, tf.float32)
image_shape = tf.cast(tf.shape(image)[:2], tf.float32)
image_height = image_shape[0]
image_width = image_shape[1]
smaller_side = tf.minimum(image_height, image_width)