1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336
| import numpy as np import numpy.random as npr
from ..fast_rcnn.config import cfg from bbox import bbox_overlaps, bbox_intersections
DEBUG = False
def generate_basic_anchors(sizes, base_size=16): base_anchor = np.array([0, 0, base_size - 1, base_size - 1], np.int32) anchors = np.zeros((len(sizes), 4), np.int32) index = 0 for h, w in sizes: anchors[index] = scale_anchor(base_anchor, h, w) index += 1 return anchors
def scale_anchor(anchor, h, w): x_ctr = (anchor[0] + anchor[2]) * 0.5 y_ctr = (anchor[1] + anchor[3]) * 0.5 scaled_anchor = anchor.copy() scaled_anchor[0] = x_ctr - w / 2 scaled_anchor[2] = x_ctr + w / 2 scaled_anchor[1] = y_ctr - h / 2 scaled_anchor[3] = y_ctr + h / 2 return scaled_anchor
def generate_anchors(base_size=16, ratios=[0.5, 1, 2], scales=2 ** np.arange(3, 6)): heights = [11, 16, 23, 33, 48, 68, 97, 139, 198, 283] widths = [16] sizes = [] for h in heights: for w in widths: sizes.append((h, w)) return generate_basic_anchors(sizes)
def bbox_transform(ex_rois, gt_rois): """ computes the distance from ground-truth boxes to the given boxes, normed by their size :param ex_rois: n * 4 numpy array, anchor boxes :param gt_rois: n * 4 numpy array, ground-truth boxes :return: deltas: n * 4 numpy array, ground-truth boxes """ ex_widths = ex_rois[:, 2] - ex_rois[:, 0] + 1.0 ex_heights = ex_rois[:, 3] - ex_rois[:, 1] + 1.0 ex_ctr_x = ex_rois[:, 0] + 0.5 * ex_widths ex_ctr_y = ex_rois[:, 1] + 0.5 * ex_heights
assert np.min(ex_widths) > 0.1 and np.min(ex_heights) > 0.1, \ 'Invalid boxes found: {} {}'. \ format(ex_rois[np.argmin(ex_widths), :], ex_rois[np.argmin(ex_heights), :])
gt_widths = gt_rois[:, 2] - gt_rois[:, 0] + 1.0 gt_heights = gt_rois[:, 3] - gt_rois[:, 1] + 1.0 gt_ctr_x = gt_rois[:, 0] + 0.5 * gt_widths gt_ctr_y = gt_rois[:, 1] + 0.5 * gt_heights
targets_dx = (gt_ctr_x - ex_ctr_x) / ex_widths targets_dy = (gt_ctr_y - ex_ctr_y) / ex_heights targets_dw = np.log(gt_widths / ex_widths) targets_dh = np.log(gt_heights / ex_heights)
targets = np.vstack( (targets_dx, targets_dy, targets_dw, targets_dh)).transpose()
return targets
def anchor_target_layer( rpn_cls_score, gt_boxes, gt_ishard, dontcare_areas, im_info, _feat_stride=[16, ], anchor_scales=[16, ]): """ Assign anchors to ground-truth targets. Produces anchor classification labels and bounding-box regression targets. Parameters ---------- rpn_cls_score: (1, H, W, Ax2) bg/fg scores of previous conv layer gt_boxes: (G, 5) vstack of [x1, y1, x2, y2, class] gt_ishard: (G, 1), 1 or 0 indicates difficult or not dontcare_areas: (D, 4), some areas may contains small objs but no labelling. D may be 0 im_info: a list of [image_height, image_width, scale_ratios] _feat_stride: the downsampling ratio of feature map to the original input image anchor_scales: the scales to the basic_anchor (basic anchor is [16, 16]) ---------- Returns ---------- rpn_labels : (HxWxA, 1), for each anchor, 0 denotes bg, 1 fg, -1 dontcare rpn_bbox_targets: (HxWxA, 4), distances of the anchors to the gt_boxes(may contains some transform) that are the regression objectives rpn_bbox_inside_weights: (HxWxA, 4) weights of each boxes, mainly accepts hyper param in cfg rpn_bbox_outside_weights: (HxWxA, 4) used to balance the fg/bg, beacuse the numbers of bgs and fgs mays significiantly different """ _anchors = generate_anchors(scales=np.array(anchor_scales)) _num_anchors = _anchors.shape[0]
_allowed_border = 0 im_info = im_info[0]
""" Algorithm: for each (H, W) location i generate 9 anchor boxes centered on cell i apply predicted bbox deltas at cell i to each of the 9 anchors filter out-of-image anchors measure GT overlap """ assert rpn_cls_score.shape[0] == 1, \ 'Only single item batches are supported'
height, width = rpn_cls_score.shape[1:3] shift_x = np.arange(0, width) * _feat_stride shift_y = np.arange(0, height) * _feat_stride shift_x, shift_y = np.meshgrid(shift_x, shift_y) shifts = np.vstack((shift_x.ravel(), shift_y.ravel(), shift_x.ravel(), shift_y.ravel())).transpose() A = _num_anchors K = shifts.shape[0] all_anchors = (_anchors.reshape((1, A, 4)) + shifts.reshape((1, K, 4)).transpose((1, 0, 2))) all_anchors = all_anchors.reshape((K * A, 4)) total_anchors = int(K * A) inds_inside = np.where( (all_anchors[:, 0] >= -_allowed_border) & (all_anchors[:, 1] >= -_allowed_border) & (all_anchors[:, 2] < im_info[1] + _allowed_border) & (all_anchors[:, 3] < im_info[0] + _allowed_border) )[0] anchors = all_anchors[inds_inside, :] labels = np.empty((len(inds_inside),), dtype=np.float32) labels.fill(-1) overlaps = bbox_overlaps( np.ascontiguousarray(anchors, dtype=np.float), np.ascontiguousarray(gt_boxes, dtype=np.float)) argmax_overlaps = overlaps.argmax(axis=1) max_overlaps = overlaps[np.arange(len(inds_inside)), argmax_overlaps] gt_argmax_overlaps = overlaps.argmax(axis=0) gt_max_overlaps = overlaps[gt_argmax_overlaps, np.arange(overlaps.shape[1])] gt_argmax_overlaps = np.where(overlaps == gt_max_overlaps)[0]
if not cfg.TRAIN.RPN_CLOBBER_POSITIVES: labels[max_overlaps < cfg.TRAIN.RPN_NEGATIVE_OVERLAP] = 0
labels[gt_argmax_overlaps] = 1 labels[max_overlaps >= cfg.TRAIN.RPN_POSITIVE_OVERLAP] = 1
if cfg.TRAIN.RPN_CLOBBER_POSITIVES: labels[max_overlaps < cfg.TRAIN.RPN_NEGATIVE_OVERLAP] = 0
if dontcare_areas is not None and dontcare_areas.shape[0] > 0: intersecs = bbox_intersections( np.ascontiguousarray(dontcare_areas, dtype=np.float), np.ascontiguousarray(anchors, dtype=np.float) ) intersecs_ = intersecs.sum(axis=0) labels[intersecs_ > cfg.TRAIN.DONTCARE_AREA_INTERSECTION_HI] = -1
if cfg.TRAIN.PRECLUDE_HARD_SAMPLES and gt_ishard is not None and gt_ishard.shape[0] > 0: assert gt_ishard.shape[0] == gt_boxes.shape[0] gt_ishard = gt_ishard.astype(int) gt_hardboxes = gt_boxes[gt_ishard == 1, :] if gt_hardboxes.shape[0] > 0: hard_overlaps = bbox_overlaps( np.ascontiguousarray(gt_hardboxes, dtype=np.float), np.ascontiguousarray(anchors, dtype=np.float)) hard_max_overlaps = hard_overlaps.max(axis=0) labels[hard_max_overlaps >= cfg.TRAIN.RPN_POSITIVE_OVERLAP] = -1 max_intersec_label_inds = hard_overlaps.argmax(axis=1) labels[max_intersec_label_inds] = -1
num_fg = int(cfg.TRAIN.RPN_FG_FRACTION * cfg.TRAIN.RPN_BATCHSIZE) 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
num_bg = cfg.TRAIN.RPN_BATCHSIZE - np.sum(labels == 1) bg_inds = np.where(labels == 0)[0] if len(bg_inds) > num_bg: disable_inds = npr.choice( bg_inds, size=(len(bg_inds) - num_bg), replace=False) labels[disable_inds] = -1
bbox_targets = np.zeros((len(inds_inside), 4), dtype=np.float32) bbox_targets = _compute_targets(anchors, gt_boxes[argmax_overlaps, :]) bbox_inside_weights = np.zeros((len(inds_inside), 4), dtype=np.float32) bbox_inside_weights[labels == 1, :] = np.array( cfg.TRAIN.RPN_BBOX_INSIDE_WEIGHTS)
bbox_outside_weights = np.zeros((len(inds_inside), 4), dtype=np.float32) if cfg.TRAIN.RPN_POSITIVE_WEIGHT < 0: positive_weights = np.ones((1, 4)) negative_weights = np.zeros((1, 4)) else: assert ((cfg.TRAIN.RPN_POSITIVE_WEIGHT > 0) & (cfg.TRAIN.RPN_POSITIVE_WEIGHT < 1)) positive_weights = (cfg.TRAIN.RPN_POSITIVE_WEIGHT / (np.sum(labels == 1)) + 1) negative_weights = ((1.0 - cfg.TRAIN.RPN_POSITIVE_WEIGHT) / (np.sum(labels == 0)) + 1) bbox_outside_weights[labels == 1, :] = positive_weights bbox_outside_weights[labels == 0, :] = negative_weights
labels = _unmap(labels, total_anchors, inds_inside, fill=-1) bbox_targets = _unmap(bbox_targets, total_anchors, inds_inside, fill=0) bbox_inside_weights = _unmap(bbox_inside_weights, total_anchors, inds_inside, fill=0) bbox_outside_weights = _unmap(bbox_outside_weights, total_anchors, inds_inside, fill=0)
labels = labels.reshape((1, height, width, A)) rpn_labels = labels
bbox_targets = bbox_targets.reshape((1, height, width, A * 4)) rpn_bbox_targets = bbox_targets
bbox_inside_weights = bbox_inside_weights.reshape((1, height, width, A * 4)) rpn_bbox_inside_weights = bbox_inside_weights
bbox_outside_weights = bbox_outside_weights.reshape((1, height, width, A * 4)) rpn_bbox_outside_weights = bbox_outside_weights rpn_data=(rpn_labels, rpn_bbox_targets, rpn_bbox_inside_weights, rpn_bbox_outside_weights) return rpn_data
def _unmap(data, count, inds, fill=0): """ Unmap a subset of item (data) back to the original set of items (of size count) """ if len(data.shape) == 1: ret = np.empty((count,), dtype=np.float32) ret.fill(fill) ret[inds] = data else: ret = np.empty((count,) + data.shape[1:], dtype=np.float32) ret.fill(fill) ret[inds, :] = data return ret
def _compute_targets(ex_rois, gt_rois): """Compute bounding-box regression targets for an image."""
assert ex_rois.shape[0] == gt_rois.shape[0] assert ex_rois.shape[1] == 4 assert gt_rois.shape[1] == 5
return bbox_transform(ex_rois, gt_rois[:, :4]).astype(np.float32, copy=False)
|