def _compute_targets(entry):
"""Compute bounding-box regression targets for an image."""
# Indices of ground-truth ROIs
rois = entry['boxes']
overlaps = entry['max_overlaps']
labels = entry['max_classes']
gt_inds = np.where((entry['gt_classes'] > 0) & (entry['is_crowd'] == 0))[0]
# Targets has format (class, tx, ty, tw, th)
targets = np.zeros((rois.shape[0], 5), dtype=np.float32)
if len(gt_inds) == 0:
# Bail if the image has no ground-truth ROIs
return targets
# Indices of examples for which we try to make predictions
ex_inds = np.where(overlaps >= cfg.TRAIN.BBOX_THRESH)[0]
# Get IoU overlap between each ex ROI and gt ROI
ex_gt_overlaps = box_utils.bbox_overlaps(
rois[ex_inds, :].astype(dtype=np.float32, copy=False),
rois[gt_inds, :].astype(dtype=np.float32, copy=False))
# Find which gt ROI each ex ROI has max overlap with:
# this will be the ex ROI's gt target
gt_assignment = ex_gt_overlaps.argmax(axis=1)
gt_rois = rois[gt_inds[gt_assignment], :]
ex_rois = rois[ex_inds, :]
# Use class "1" for all boxes if using class_agnostic_bbox_reg
targets[ex_inds,
0] = (1 if cfg.MODEL.CLS_AGNOSTIC_BBOX_REG else labels[ex_inds])
targets[ex_inds,
1:] = box_utils.bbox_transform_inv(ex_rois, gt_rois,
cfg.MODEL.BBOX_REG_WEIGHTS)
return targets
def _merge_proposal_boxes_into_roidb(roidb, box_list):
"""Add proposal boxes to each roidb entry."""
assert len(box_list) == len(roidb)
for i, entry in enumerate(roidb):
boxes = box_list[i]
num_boxes = boxes.shape[0]
gt_overlaps = np.zeros((num_boxes, entry['gt_overlaps'].shape[1]),
dtype=entry['gt_overlaps'].dtype)
box_to_gt_ind_map = -np.ones(
(num_boxes), dtype=entry['box_to_gt_ind_map'].dtype)
# Note: unlike in other places, here we intentionally include all gt
# rois, even ones marked as crowd. Boxes that overlap with crowds will
# be filtered out later (see: _filter_crowd_proposals).
gt_inds = np.where(entry['gt_classes'] > 0)[0]
if len(gt_inds) > 0:
gt_boxes = entry['boxes'][gt_inds, :]
gt_classes = entry['gt_classes'][gt_inds]
proposal_to_gt_overlaps = box_utils.bbox_overlaps(
boxes.astype(dtype=np.float32, copy=False),
gt_boxes.astype(dtype=np.float32, copy=False))
# Gt box that overlaps each input box the most
# (ties are broken arbitrarily by class order)
argmaxes = proposal_to_gt_overlaps.argmax(axis=1)
# Amount of that overlap
maxes = proposal_to_gt_overlaps.max(axis=1)
# Those boxes with non-zero overlap with gt boxes
I = np.where(maxes > 0)[0]
# Record max overlaps with the class of the appropriate gt box
gt_overlaps[I, gt_classes[argmaxes[I]]] = maxes[I]
box_to_gt_ind_map[I] = gt_inds[argmaxes[I]]
entry['boxes'] = np.append(entry['boxes'],
boxes.astype(entry['boxes'].dtype,
copy=False),
axis=0)
entry['gt_classes'] = np.append(
entry['gt_classes'],
np.zeros((num_boxes), dtype=entry['gt_classes'].dtype))
entry['seg_areas'] = np.append(
entry['seg_areas'],
np.zeros((num_boxes), dtype=entry['seg_areas'].dtype))
entry['gt_overlaps'] = np.append(entry['gt_overlaps'].toarray(),
gt_overlaps,
axis=0)
entry['gt_overlaps'] = scipy.sparse.csr_matrix(entry['gt_overlaps'])
entry['is_crowd'] = np.append(
entry['is_crowd'],
np.zeros((num_boxes), dtype=entry['is_crowd'].dtype))
entry['box_to_gt_ind_map'] = np.append(
entry['box_to_gt_ind_map'],
box_to_gt_ind_map.astype(entry['box_to_gt_ind_map'].dtype,
copy=False))
def add_uv_rcnn_blobs(blobs, sampled_boxes, roidb, im_scale, batch_idx):
M = cfg.UVRCNN.HEATMAP_SIZE
IsFlipped = roidb['flipped']
#
polys_gt_inds = np.where(roidb['ignore_UV_body'] == 0)[0]
boxes_from_polys = [roidb['boxes'][i, :] for i in polys_gt_inds]
if not (boxes_from_polys):
pass
else:
boxes_from_polys = np.vstack(boxes_from_polys)
boxes_from_polys = np.array(boxes_from_polys)
fg_inds = np.where(blobs['labels_int32'] > 0)[0]
roi_has_mask = np.zeros(blobs['labels_int32'].shape)
if (bool(boxes_from_polys.any()) & (fg_inds.shape[0] > 0)):
rois_fg = sampled_boxes[fg_inds]
#
rois_fg.astype(np.float32, copy=False)
boxes_from_polys.astype(np.float32, copy=False)
#
overlaps_bbfg_bbpolys = box_utils.bbox_overlaps(
rois_fg.astype(np.float32, copy=False),
boxes_from_polys.astype(np.float32, copy=False))
fg_polys_value = np.max(overlaps_bbfg_bbpolys, axis=1)
fg_inds = fg_inds[fg_polys_value > 0.7]
if (bool(boxes_from_polys.any()) & (fg_inds.shape[0] > 0)):
for jj in fg_inds:
roi_has_mask[jj] = 1
# Create blobs for densepose supervision.
################################################## The mask
All_labels = blob_utils.zeros((fg_inds.shape[0], M**2), int32=True)
All_Weights = blob_utils.zeros((fg_inds.shape[0], M**2), int32=True)
################################################# The points
X_points = blob_utils.zeros((fg_inds.shape[0], 196), int32=False)
Y_points = blob_utils.zeros((fg_inds.shape[0], 196), int32=False)
Ind_points = blob_utils.zeros((fg_inds.shape[0], 196), int32=True)
I_points = blob_utils.zeros((fg_inds.shape[0], 196), int32=True)
U_points = blob_utils.zeros((fg_inds.shape[0], 196), int32=False)
V_points = blob_utils.zeros((fg_inds.shape[0], 196), int32=False)
Uv_point_weights = blob_utils.zeros((fg_inds.shape[0], 196),
int32=False)
#################################################
rois_fg = sampled_boxes[fg_inds]
overlaps_bbfg_bbpolys = box_utils.bbox_overlaps(
rois_fg.astype(np.float32, copy=False),
boxes_from_polys.astype(np.float32, copy=False))
fg_polys_inds = np.argmax(overlaps_bbfg_bbpolys, axis=1)
for i in range(rois_fg.shape[0]):
#
fg_polys_ind = polys_gt_inds[fg_polys_inds[i]]
#
Ilabel = segm_utils.GetDensePoseMask(
roidb['dp_masks'][fg_polys_ind])
#
GT_I = np.array(roidb['dp_I'][fg_polys_ind])
GT_U = np.array(roidb['dp_U'][fg_polys_ind])
GT_V = np.array(roidb['dp_V'][fg_polys_ind])
GT_x = np.array(roidb['dp_x'][fg_polys_ind])
GT_y = np.array(roidb['dp_y'][fg_polys_ind])
GT_weights = np.ones(GT_I.shape).astype(np.float32)
#
## Do the flipping of the densepose annotation !
if (IsFlipped):
GT_I, GT_U, GT_V, GT_x, GT_y, Ilabel = DP.get_symmetric_densepose(
GT_I, GT_U, GT_V, GT_x, GT_y, Ilabel)
#
roi_fg = rois_fg[i]
roi_gt = boxes_from_polys[fg_polys_inds[i], :]
#
x1 = roi_fg[0]
x2 = roi_fg[2]
y1 = roi_fg[1]
y2 = roi_fg[3]
#
x1_source = roi_gt[0]
x2_source = roi_gt[2]
y1_source = roi_gt[1]
y2_source = roi_gt[3]
#
x_targets = (np.arange(x1, x2, (x2 - x1) / M) -
x1_source) * (256. / (x2_source - x1_source))
y_targets = (np.arange(y1, y2, (y2 - y1) / M) -
y1_source) * (256. / (y2_source - y1_source))
#
x_targets = x_targets[
0:
M] ## Strangely sometimes it can be M+1, so make sure size is OK!
y_targets = y_targets[0:M]
#
[X_targets, Y_targets] = np.meshgrid(x_targets, y_targets)
New_Index = cv2.remap(Ilabel,
X_targets.astype(np.float32),
Y_targets.astype(np.float32),
interpolation=cv2.INTER_NEAREST,
borderMode=cv2.BORDER_CONSTANT,
borderValue=(0))
#
All_L = np.zeros(New_Index.shape)
All_W = np.ones(New_Index.shape)
#
All_L = New_Index
#
gt_length_x = x2_source - x1_source
gt_length_y = y2_source - y1_source
#
GT_y = ((GT_y / 256. * gt_length_y) + y1_source - y1) * (M /
(y2 - y1))
GT_x = ((GT_x / 256. * gt_length_x) + x1_source - x1) * (M /
(x2 - x1))
#
GT_I[GT_y < 0] = 0
GT_I[GT_y > (M - 1)] = 0
GT_I[GT_x < 0] = 0
GT_I[GT_x > (M - 1)] = 0
#
points_inside = GT_I > 0
GT_U = GT_U[points_inside]
GT_V = GT_V[points_inside]
GT_x = GT_x[points_inside]
GT_y = GT_y[points_inside]
GT_weights = GT_weights[points_inside]
GT_I = GT_I[points_inside]
#
X_points[i, 0:len(GT_x)] = GT_x
Y_points[i, 0:len(GT_y)] = GT_y
Ind_points[i, 0:len(GT_I)] = i
I_points[i, 0:len(GT_I)] = GT_I
U_points[i, 0:len(GT_U)] = GT_U
V_points[i, 0:len(GT_V)] = GT_V
Uv_point_weights[i, 0:len(GT_weights)] = GT_weights
#
All_labels[i, :] = np.reshape(All_L.astype(np.int32), M**2)
All_Weights[i, :] = np.reshape(All_W.astype(np.int32), M**2)
##
else:
bg_inds = np.where(blobs['labels_int32'] == 0)[0]
#
if (len(bg_inds) == 0):
rois_fg = sampled_boxes[0].reshape((1, -1))
else:
rois_fg = sampled_boxes[bg_inds[0]].reshape((1, -1))
roi_has_mask[0] = 1
#
X_points = blob_utils.zeros((1, 196), int32=False)
Y_points = blob_utils.zeros((1, 196), int32=False)
Ind_points = blob_utils.zeros((1, 196), int32=True)
I_points = blob_utils.zeros((1, 196), int32=True)
U_points = blob_utils.zeros((1, 196), int32=False)
V_points = blob_utils.zeros((1, 196), int32=False)
Uv_point_weights = blob_utils.zeros((1, 196), int32=False)
#
All_labels = -blob_utils.ones((1, M**2), int32=True) * 0 ## zeros
All_Weights = -blob_utils.ones((1, M**2), int32=True) * 0 ## zeros
#
rois_fg *= im_scale
repeated_batch_idx = batch_idx * blob_utils.ones((rois_fg.shape[0], 1))
rois_fg = np.hstack((repeated_batch_idx, rois_fg))
#
K = cfg.UVRCNN.NUM_PATCHES
#
U_points = np.tile(U_points, [1, K + 1])
V_points = np.tile(V_points, [1, K + 1])
Uv_Weight_Points = np.zeros(U_points.shape)
#
for jjj in range(1, K + 1):
Uv_Weight_Points[:, jjj * I_points.shape[1]:(jjj + 1) *
I_points.shape[1]] = (I_points == jjj).astype(
np.float32)
#
################
# Update blobs dict with Mask R-CNN blobs
###############
#
blobs['uv_rois'] = np.array(rois_fg)
blobs['roi_has_uv_int32'] = np.array(roi_has_mask).astype(np.int32)
##
blobs['uv_ann_labels'] = np.array(All_labels).astype(np.int32)
blobs['uv_ann_weights'] = np.array(All_Weights).astype(np.float32)
#
##########################
blobs['uv_X_points'] = X_points.astype(np.float32)
blobs['uv_Y_points'] = Y_points.astype(np.float32)
blobs['uv_Ind_points'] = Ind_points.astype(np.float32)
blobs['uv_I_points'] = I_points.astype(np.float32)
blobs['uv_U_points'] = U_points.astype(
np.float32) #### VERY IMPORTANT : These are switched here :
blobs['uv_V_points'] = V_points.astype(np.float32)
blobs['uv_point_weights'] = Uv_Weight_Points.astype(np.float32)
def add_parsing_rcnn_blobs(blobs, sampled_boxes, roidb, im_scale, batch_idx):
"""Add parsing R-CNN specific blobs to the input blob dictionary."""
# Prepare the parsing targets by associating one gt parsing to each training roi
# that has a fg (non-bg) class label.
M = cfg.PRCNN.RESOLUTION
polys_gt_inds = np.where((roidb['gt_classes'] > 0)
& (roidb['is_crowd'] == 0))[0]
parsing_gt = [roidb['parsing'][i] for i in polys_gt_inds]
boxes_from_png = parsing_utils.parsing_to_boxes(parsing_gt,
roidb['flipped'])
fg_inds = np.where(blobs['labels_int32'] > 0)[0]
if fg_inds.shape[0] > 0:
if cfg.PRCNN.ROI_BATCH_SIZE > 0:
fg_rois_per_this_image = np.minimum(cfg.PRCNN.ROI_BATCH_SIZE,
fg_inds.shape[0])
fg_inds = npr.choice(fg_inds,
size=fg_rois_per_this_image,
replace=False)
parsings = blob_utils.zeros((fg_inds.shape[0], M**2), int32=True)
# Find overlap between all foreground rois and the bounding boxes
# enclosing each segmentation
rois_fg = sampled_boxes[fg_inds]
overlaps_bbfg_bbpolys = box_utils.bbox_overlaps(
rois_fg.astype(np.float32, copy=False),
boxes_from_png.astype(np.float32, copy=False))
# Map from each fg rois to the index of the parsing with highest overlap
# (measured by bbox overlap)
fg_polys_inds = np.argmax(overlaps_bbfg_bbpolys, axis=1)
# add fg targets
for i in range(rois_fg.shape[0]):
fg_polys_ind = fg_polys_inds[i]
parsing_gt_fg = parsing_gt[fg_polys_ind]
roi_fg = rois_fg[i]
# Rasterize the portion of the polygon mask within the given fg roi
# to an M x M binary image
parsing = parsing_utils.parsing_wrt_box(parsing_gt_fg, roi_fg, M,
roidb['flipped'])
parsings[i, :] = parsing
weights = blob_utils.ones((rois_fg.shape[0], M**2))
else: # If there are no fg masks (it does happen)
# The network cannot handle empty blobs, so we must provide a mask
# We simply take the first bg roi, given it an all -1's mask (ignore
# label), and label it with class zero (bg).
bg_inds = np.where(blobs['labels_int32'] == 0)[0]
# rois_fg is actually one background roi, but that's ok because ...
if (len(bg_inds) == 0):
rois_fg = sampled_boxes[0].reshape((1, -1))
else:
rois_fg = sampled_boxes[bg_inds[0]].reshape((1, -1))
# We give it an -1's blob (ignore label)
parsings = blob_utils.zeros((1, M**2), int32=True)
# Mark that the first roi has a mask
weights = blob_utils.zeros((1, M**2))
parsings = np.reshape(parsings, (-1, 1))
weights = np.reshape(weights, (-1, 1))
# Scale rois_fg and format as (batch_idx, x1, y1, x2, y2)
rois_fg *= im_scale
repeated_batch_idx = batch_idx * blob_utils.ones((rois_fg.shape[0], 1))
rois_fg = np.hstack((repeated_batch_idx, rois_fg))
# Update blobs dict with Mask R-CNN blobs
blobs['parsing_rois'] = rois_fg
blobs['parsing_weights'] = weights
blobs['parsing_int32'] = parsings
def add_mask_rcnn_blobs(blobs, sampled_boxes, roidb, im_scale, batch_idx):
"""Add Mask R-CNN specific blobs to the input blob dictionary."""
# Prepare the mask targets by associating one gt mask to each training roi
# that has a fg (non-bg) class label.
M = cfg.MRCNN.RESOLUTION
polys_gt_inds = np.where((roidb['gt_classes'] > 0)
& (roidb['is_crowd'] == 0))[0]
polys_gt = [roidb['segms'][i] for i in polys_gt_inds]
boxes_from_polys = segm_utils.polys_to_boxes(polys_gt)
# boxes_from_polys = [roidb['boxes'][i] for i in polys_gt_inds]
fg_inds = np.where(blobs['labels_int32'] > 0)[0]
roi_has_mask = blobs['labels_int32'].copy()
roi_has_mask[roi_has_mask > 0] = 1
if fg_inds.shape[0] > 0:
if cfg.MRCNN.ROI_BATCH_SIZE > 0:
fg_rois_per_this_image = np.minimum(cfg.MRCNN.ROI_BATCH_SIZE,
fg_inds.shape[0])
fg_inds = npr.choice(fg_inds,
size=fg_rois_per_this_image,
replace=False)
# Class labels for the foreground rois
mask_class_labels = blobs['labels_int32'][fg_inds]
masks = blob_utils.zeros((fg_inds.shape[0], M**2), int32=True)
# Find overlap between all foreground rois and the bounding boxes
# enclosing each segmentation
rois_fg = sampled_boxes[fg_inds]
overlaps_bbfg_bbpolys = box_utils.bbox_overlaps(
rois_fg.astype(np.float32, copy=False),
boxes_from_polys.astype(np.float32, copy=False))
# Map from each fg rois to the index of the mask with highest overlap
# (measured by bbox overlap)
fg_polys_inds = np.argmax(overlaps_bbfg_bbpolys, axis=1)
# add fg targets
for i in range(rois_fg.shape[0]):
fg_polys_ind = fg_polys_inds[i]
poly_gt = polys_gt[fg_polys_ind]
roi_fg = rois_fg[i]
# Rasterize the portion of the polygon mask within the given fg roi
# to an M x M binary image
mask = segm_utils.polys_to_mask_wrt_box(poly_gt, roi_fg, M)
mask = np.array(mask > 0, dtype=np.int32) # Ensure it's binary
masks[i, :] = np.reshape(mask, M**2)
else: # If there are no fg masks (it does happen)
# The network cannot handle empty blobs, so we must provide a mask
# We simply take the first bg roi, given it an all -1's mask (ignore
# label), and label it with class zero (bg).
bg_inds = np.where(blobs['labels_int32'] == 0)[0]
# rois_fg is actually one background roi, but that's ok because ...
rois_fg = sampled_boxes[bg_inds[0]].reshape((1, -1))
# We give it an -1's blob (ignore label)
masks = -blob_utils.ones((1, M**2), int32=True)
# We label it with class = 0 (background)
mask_class_labels = blob_utils.zeros((1, ))
# Mark that the first roi has a mask
roi_has_mask[0] = 1
if cfg.MRCNN.CLS_SPECIFIC_MASK:
masks = _expand_to_class_specific_mask_targets(masks,
mask_class_labels)
# Scale rois_fg and format as (batch_idx, x1, y1, x2, y2)
rois_fg *= im_scale
repeated_batch_idx = batch_idx * blob_utils.ones((rois_fg.shape[0], 1))
rois_fg = np.hstack((repeated_batch_idx, rois_fg))
# Update blobs dict with Mask R-CNN blobs
blobs['mask_rois'] = rois_fg
blobs['roi_has_mask_int32'] = roi_has_mask
blobs['masks_int32'] = masks
def _get_retinanet_blobs(foas, all_anchors, gt_boxes, gt_classes, im_width,
im_height):
total_anchors = all_anchors.shape[0]
logger.debug('Getting mad blobs: im_height {} im_width: {}'.format(
im_height, im_width))
inds_inside = np.arange(all_anchors.shape[0]) # 0, 1... 371349
anchors = all_anchors
num_inside = len(inds_inside) # 371349
logger.debug('total_anchors: {}'.format(total_anchors))
logger.debug('inds_inside: {}'.format(num_inside))
logger.debug('anchors.shape: {}'.format(anchors.shape))
# Compute anchor labels:
# label=1 is positive, 0 is negative, -1 is don't care (ignore)
labels = np.empty((num_inside, ), dtype=np.float32)
labels.fill(-1)
if len(gt_boxes) > 0:
# Compute overlaps between the anchors and the gt boxes overlaps
anchor_by_gt_overlap = box_utils.bbox_overlaps(
anchors, gt_boxes) # (371349, 17)
# Map from anchor to gt box that has highest overlap
anchor_to_gt_argmax = anchor_by_gt_overlap.argmax(
axis=1) # (371349,) this is index
# For each anchor, amount of overlap with most overlapping gt box
anchor_to_gt_max = anchor_by_gt_overlap[ # (371349,) this is area
np.arange(num_inside), anchor_to_gt_argmax]
# Map from gt box to an anchor that has highest overlap
gt_to_anchor_argmax = anchor_by_gt_overlap.argmax(
axis=0) # (17,) index
# For each gt box, amount of overlap with most overlapping anchor
gt_to_anchor_max = anchor_by_gt_overlap[ # (17,) area
gt_to_anchor_argmax,
np.arange(anchor_by_gt_overlap.shape[1])]
# Find all anchors that share the max overlap amount
# (this includes many ties)
anchors_with_max_overlap = np.where( # (21,) find all anchors with most overlaps
anchor_by_gt_overlap == gt_to_anchor_max)[0]
# Fg label: for each gt use anchors with highest overlap
# (including ties)
gt_inds = anchor_to_gt_argmax[anchors_with_max_overlap] # 416
labels[anchors_with_max_overlap] = gt_classes[gt_inds]
# Fg label: above threshold IOU
inds = anchor_to_gt_max >= cfg.RETINANET.POSITIVE_OVERLAP
gt_inds = anchor_to_gt_argmax[inds]
labels[inds] = gt_classes[
gt_inds] # for all anchors, inds are valued by gt_inds, this gives class values 1~80
fg_inds = np.where(labels >= 1)[0]
bg_inds = np.where(anchor_to_gt_max < cfg.RETINANET.NEGATIVE_OVERLAP)[0]
labels[bg_inds] = 0
num_fg, num_bg = len(fg_inds), len(bg_inds)
bbox_targets = np.zeros((num_inside, 4), dtype=np.float32)
bbox_targets[fg_inds, :] = data_utils.compute_targets(
anchors[fg_inds, :], gt_boxes[anchor_to_gt_argmax[fg_inds], :])
# Map up to original set of anchors
labels = data_utils.unmap(labels, total_anchors, inds_inside, fill=-1)
bbox_targets = data_utils.unmap(bbox_targets,
total_anchors,
inds_inside,
fill=0)
# Split the generated labels, etc. into labels per each field of anchors
blobs_out = []
start_idx = 0
for foa in foas:
H = foa.field_size
W = foa.field_size
end_idx = start_idx + H * W
_labels = labels[start_idx:end_idx]
triangle_start_idx = start_idx
_bbox_targets = bbox_targets[start_idx:end_idx, :]
start_idx = end_idx
# labels output with shape (1, height, width)
_labels = _labels.reshape((1, 1, H, W))
# bbox_targets output with shape (1, 4 * A, height, width)
_bbox_targets = _bbox_targets.reshape(
(1, H, W, 4)).transpose(0, 3, 1, 2)
stride = foa.stride
w = int(im_width / stride)
h = int(im_height / stride)
# data for select_smooth_l1 loss
num_classes = cfg.MODEL.NUM_CLASSES - 1
inds_4d = np.where(_labels > 0)
M = len(inds_4d)
_roi_bbox_targets = np.zeros((0, 4))
_roi_fg_bbox_locs = np.zeros((0, 4))
if M > 0:
im_inds, y, x = inds_4d[0], inds_4d[2], inds_4d[3]
_roi_bbox_targets = np.zeros((len(im_inds), 4))
_roi_fg_bbox_locs = np.zeros((len(im_inds), 4))
lbls = _labels[im_inds, :, y, x]
for i, lbl in enumerate(lbls):
l = lbl[0] - 1
if not cfg.RETINANET.CLASS_SPECIFIC_BBOX:
l = 0
assert l >= 0 and l < num_classes, 'label out of the range'
_roi_bbox_targets[i, :] = _bbox_targets[:, :, y[i], x[i]]
_roi_fg_bbox_locs[i, :] = np.array([[0, l, y[i], x[i]]])
if _roi_bbox_targets.astype(
np.float32).shape[0] == 0 and _roi_fg_bbox_locs.astype(
np.float32).shape[0] == 0:
blobs_out.append(
dict(
retnet_cls_labels=_labels[:, :, 0:h, 0:w].astype(np.int32),
retnet_roi_bbox_targets=_roi_bbox_targets.astype(
np.float32),
retnet_roi_fg_bbox_locs=_roi_fg_bbox_locs.astype(
np.float32),
# retnet_roi_bbox_targets=np.array([[0, 0, 0, 0]]),
# retnet_roi_fg_bbox_locs=np.array([[0, 0, 0, 0]]),
))
# we don't add zero padding here, because this is inside the loop of foa, we don't
# want every anchor to have padding, instead we want to firstly sum all anchors in a FPN of an image, and then check if it's emtpy
else:
blobs_out.append(
dict(
retnet_cls_labels=_labels[:, :, 0:h, 0:w].astype(np.int32),
retnet_roi_bbox_targets=_roi_bbox_targets.astype(
np.float32),
retnet_roi_fg_bbox_locs=_roi_fg_bbox_locs.astype(
np.float32),
))
out_num_fg = np.array([num_fg + 1.0], dtype=np.float32)
out_num_bg = (np.array([num_bg + 1.0]) * (cfg.MODEL.NUM_CLASSES - 1) +
out_num_fg * (cfg.MODEL.NUM_CLASSES - 2))
return blobs_out, out_num_fg, out_num_bg
def evaluate_box_proposals(json_dataset,
roidb,
thresholds=None,
area='all',
limit=None):
"""Evaluate detection proposal recall metrics. This function is a much
faster alternative to the official COCO API recall evaluation code. However,
it produces slightly different results.
"""
# Record max overlap value for each gt box
# Return vector of overlap values
areas = {
'all': 0,
'small': 1,
'medium': 2,
'large': 3,
'96-128': 4,
'128-256': 5,
'256-512': 6,
'512-inf': 7
}
area_ranges = [
[0**2, 1e5**2], # all
[0**2, 32**2], # small
[32**2, 96**2], # medium
[96**2, 1e5**2], # large
[96**2, 128**2], # 96-128
[128**2, 256**2], # 128-256
[256**2, 512**2], # 256-512
[512**2, 1e5**2]
] # 512-inf
assert area in areas, 'Unknown area range: {}'.format(area)
area_range = area_ranges[areas[area]]
gt_overlaps = np.zeros(0)
num_pos = 0
for entry in roidb:
gt_inds = np.where((entry['gt_classes'] > 0)
& (entry['is_crowd'] == 0))[0]
gt_boxes = entry['boxes'][gt_inds, :]
gt_areas = entry['seg_areas'][gt_inds]
valid_gt_inds = np.where((gt_areas >= area_range[0])
& (gt_areas <= area_range[1]))[0]
gt_boxes = gt_boxes[valid_gt_inds, :]
num_pos += len(valid_gt_inds)
non_gt_inds = np.where(entry['gt_classes'] == 0)[0]
boxes = entry['boxes'][non_gt_inds, :]
if boxes.shape[0] == 0:
continue
if limit is not None and boxes.shape[0] > limit:
boxes = boxes[:limit, :]
overlaps = box_utils.bbox_overlaps(
boxes.astype(dtype=np.float32, copy=False),
gt_boxes.astype(dtype=np.float32, copy=False))
_gt_overlaps = np.zeros((gt_boxes.shape[0]))
for j in range(min(boxes.shape[0], gt_boxes.shape[0])):
# find which proposal box maximally covers each gt box
argmax_overlaps = overlaps.argmax(axis=0)
# and get the iou amount of coverage for each gt box
max_overlaps = overlaps.max(axis=0)
# find which gt box is 'best' covered (i.e. 'best' = most iou)
gt_ind = max_overlaps.argmax()
gt_ovr = max_overlaps.max()
assert gt_ovr >= 0
# find the proposal box that covers the best covered gt box
box_ind = argmax_overlaps[gt_ind]
# record the iou coverage of this gt box
_gt_overlaps[j] = overlaps[box_ind, gt_ind]
assert _gt_overlaps[j] == gt_ovr
# mark the proposal box and the gt box as used
overlaps[box_ind, :] = -1
overlaps[:, gt_ind] = -1
# append recorded iou coverage level
gt_overlaps = np.hstack((gt_overlaps, _gt_overlaps))
gt_overlaps = np.sort(gt_overlaps)
if thresholds is None:
step = 0.05
thresholds = np.arange(0.5, 0.95 + 1e-5, step)
recalls = np.zeros_like(thresholds)
# compute recall for each iou threshold
for i, t in enumerate(thresholds):
recalls[i] = (gt_overlaps >= t).sum() / float(num_pos)
# ar = 2 * np.trapz(recalls, thresholds)
ar = recalls.mean()
return {
'ar': ar,
'recalls': recalls,
'thresholds': thresholds,
'gt_overlaps': gt_overlaps,
'num_pos': num_pos
}
def _get_rpn_blobs(im_height, im_width, foas, all_anchors, gt_boxes):
total_anchors = all_anchors.shape[0]
straddle_thresh = cfg.TRAIN.RPN_STRADDLE_THRESH
if straddle_thresh >= 0:
# Only keep anchors inside the image by a margin of straddle_thresh
# Set TRAIN.RPN_STRADDLE_THRESH to -1 (or a large value) to keep all
# anchors
inds_inside = np.where(
(all_anchors[:, 0] >= -straddle_thresh)
& (all_anchors[:, 1] >= -straddle_thresh)
& (all_anchors[:, 2] < im_width + straddle_thresh)
& (all_anchors[:, 3] < im_height + straddle_thresh))[0]
# keep only inside anchors
anchors = all_anchors[inds_inside, :]
else:
inds_inside = np.arange(all_anchors.shape[0])
anchors = all_anchors
num_inside = len(inds_inside)
logger.debug('total_anchors: %d', total_anchors)
logger.debug('inds_inside: %d', num_inside)
logger.debug('anchors.shape: %s', str(anchors.shape))
# Compute anchor labels:
# label=1 is positive, 0 is negative, -1 is don't care (ignore)
labels = np.empty((num_inside, ), dtype=np.int32)
labels.fill(-1)
if len(gt_boxes) > 0:
# Compute overlaps between the anchors and the gt boxes overlaps
anchor_by_gt_overlap = box_utils.bbox_overlaps(anchors, gt_boxes)
# Map from anchor to gt box that has highest overlap
anchor_to_gt_argmax = anchor_by_gt_overlap.argmax(axis=1)
# For each anchor, amount of overlap with most overlapping gt box
anchor_to_gt_max = anchor_by_gt_overlap[np.arange(num_inside),
anchor_to_gt_argmax]
# Map from gt box to an anchor that has highest overlap
gt_to_anchor_argmax = anchor_by_gt_overlap.argmax(axis=0)
# For each gt box, amount of overlap with most overlapping anchor
gt_to_anchor_max = anchor_by_gt_overlap[
gt_to_anchor_argmax,
np.arange(anchor_by_gt_overlap.shape[1])]
# Find all anchors that share the max overlap amount
# (this includes many ties)
anchors_with_max_overlap = np.where(
anchor_by_gt_overlap == gt_to_anchor_max)[0]
# Fg label: for each gt use anchors with highest overlap
# (including ties)
labels[anchors_with_max_overlap] = 1
# Fg label: above threshold IOU
labels[anchor_to_gt_max >= cfg.TRAIN.RPN_POSITIVE_OVERLAP] = 1
# subsample positive labels if we have too many
num_fg = int(cfg.TRAIN.RPN_FG_FRACTION * cfg.TRAIN.RPN_BATCH_SIZE_PER_IM)
fg_inds = np.where(labels == 1)[0]
if len(fg_inds) > num_fg:
disable_inds = npr.choice(fg_inds,
size=(len(fg_inds) - num_fg),
replace=False)
labels[disable_inds] = -1
fg_inds = np.where(labels == 1)[0]
# subsample negative labels if we have too many
# (samples with replacement, but since the set of bg inds is large most
# samples will not have repeats)
num_bg = cfg.TRAIN.RPN_BATCH_SIZE_PER_IM - np.sum(labels == 1)
bg_inds = np.where(anchor_to_gt_max < cfg.TRAIN.RPN_NEGATIVE_OVERLAP)[0]
if len(bg_inds) > num_bg:
enable_inds = bg_inds[npr.randint(len(bg_inds), size=num_bg)]
labels[enable_inds] = 0
bg_inds = np.where(labels == 0)[0]
bbox_targets = np.zeros((num_inside, 4), dtype=np.float32)
bbox_targets[fg_inds, :] = data_utils.compute_targets(
anchors[fg_inds, :], gt_boxes[anchor_to_gt_argmax[fg_inds], :])
# Bbox regression loss has the form:
# loss(x) = weight_outside * L(weight_inside * x)
# Inside weights allow us to set zero loss on an element-wise basis
# Bbox regression is only trained on positive examples so we set their
# weights to 1.0 (or otherwise if config is different) and 0 otherwise
bbox_inside_weights = np.zeros((num_inside, 4), dtype=np.float32)
bbox_inside_weights[labels == 1, :] = (1.0, 1.0, 1.0, 1.0)
# The bbox regression loss only averages by the number of images in the
# mini-batch, whereas we need to average by the total number of example
# anchors selected
# Outside weights are used to scale each element-wise loss so the final
# average over the mini-batch is correct
bbox_outside_weights = np.zeros((num_inside, 4), dtype=np.float32)
# uniform weighting of examples (given non-uniform sampling)
num_examples = np.sum(labels >= 0)
bbox_outside_weights[labels == 1, :] = 1.0 / num_examples
bbox_outside_weights[labels == 0, :] = 1.0 / num_examples
# Map up to original set of anchors
labels = data_utils.unmap(labels, total_anchors, inds_inside, fill=-1)
bbox_targets = data_utils.unmap(bbox_targets,
total_anchors,
inds_inside,
fill=0)
bbox_inside_weights = data_utils.unmap(bbox_inside_weights,
total_anchors,
inds_inside,
fill=0)
bbox_outside_weights = data_utils.unmap(bbox_outside_weights,
total_anchors,
inds_inside,
fill=0)
# Split the generated labels, etc. into labels per each field of anchors
blobs_out = []
start_idx = 0
for foa in foas:
H = foa.field_size
W = foa.field_size
A = foa.num_cell_anchors
end_idx = start_idx + H * W * A
_labels = labels[start_idx:end_idx]
_bbox_targets = bbox_targets[start_idx:end_idx, :]
_bbox_inside_weights = bbox_inside_weights[start_idx:end_idx, :]
_bbox_outside_weights = bbox_outside_weights[start_idx:end_idx, :]
start_idx = end_idx
# labels output with shape (1, A, height, width)
_labels = _labels.reshape((1, H, W, A)).transpose(0, 3, 1, 2)
# bbox_targets output with shape (1, 4 * A, height, width)
_bbox_targets = _bbox_targets.reshape(
(1, H, W, A * 4)).transpose(0, 3, 1, 2)
# bbox_inside_weights output with shape (1, 4 * A, height, width)
_bbox_inside_weights = _bbox_inside_weights.reshape(
(1, H, W, A * 4)).transpose(0, 3, 1, 2)
# bbox_outside_weights output with shape (1, 4 * A, height, width)
_bbox_outside_weights = _bbox_outside_weights.reshape(
(1, H, W, A * 4)).transpose(0, 3, 1, 2)
blobs_out.append(
dict(rpn_labels_int32_wide=_labels,
rpn_bbox_targets_wide=_bbox_targets,
rpn_bbox_inside_weights_wide=_bbox_inside_weights,
rpn_bbox_outside_weights_wide=_bbox_outside_weights))
return blobs_out[0] if len(blobs_out) == 1 else blobs_out