def im_detect_bbox_aug(model, images, device):
# Collect detections computed under different transformations
boxlists_ts = []
for _ in range(len(images)):
boxlists_ts.append([])
def add_preds_t(boxlists_t):
for i, boxlist_t in enumerate(boxlists_t):
if len(boxlists_ts[i]) == 0:
# The first one is identity transform, no need to resize the boxlist
boxlists_ts[i].append(boxlist_t)
else:
# Resize the boxlist as the first one
boxlists_ts[i].append(boxlist_t.resize(boxlists_ts[i][0].size))
# Compute detections for the original image (identity transform)
boxlists_i = im_detect_bbox(
model, images, cfg.INPUT.MIN_SIZE_TEST, cfg.INPUT.MAX_SIZE_TEST, device
)
add_preds_t(boxlists_i)
# Perform detection on the horizontally flipped image
if cfg.TEST.BBOX_AUG.H_FLIP:
boxlists_hf = im_detect_bbox_hflip(
model, images, cfg.INPUT.MIN_SIZE_TEST, cfg.INPUT.MAX_SIZE_TEST, device
)
add_preds_t(boxlists_hf)
# Compute detections at different scales
for scale in cfg.TEST.BBOX_AUG.SCALES:
max_size = cfg.TEST.BBOX_AUG.MAX_SIZE
boxlists_scl = im_detect_bbox_scale(
model, images, scale, max_size, device
)
add_preds_t(boxlists_scl)
if cfg.TEST.BBOX_AUG.SCALE_H_FLIP:
boxlists_scl_hf = im_detect_bbox_scale(
model, images, scale, max_size, device, hflip=True
)
add_preds_t(boxlists_scl_hf)
# Merge boxlists detected by different bbox aug params
boxlists = []
for i, boxlist_ts in enumerate(boxlists_ts):
bbox = torch.cat([boxlist_t.bbox for boxlist_t in boxlist_ts])
scores = torch.cat([boxlist_t.get_field('scores') for boxlist_t in boxlist_ts])
boxlist = BoxList(bbox, boxlist_ts[0].size, boxlist_ts[0].mode)
boxlist.add_field('scores', scores)
boxlists.append(boxlist)
# Apply NMS and limit the final detections
results = []
post_processor = make_roi_box_post_processor(cfg)
for boxlist in boxlists:
results.append(post_processor.filter_results(boxlist, cfg.MODEL.ROI_BOX_HEAD.NUM_CLASSES))
return results
def __getitem__(self, idx):
img, anno = super(COCODataset, self).__getitem__(idx)
# filter crowd annotations
# TODO might be better to add an extra field
anno = [obj for obj in anno if obj["iscrowd"] == 0]
boxes = [obj["bbox"] for obj in anno]
boxes = torch.as_tensor(boxes).reshape(-1, 4) # guard against no boxes
target = BoxList(boxes, img.size, mode="xywh").convert("xyxy")
classes = [obj["category_id"] for obj in anno]
classes = [self.json_category_id_to_contiguous_id[c] for c in classes]
classes = torch.tensor(classes)
target.add_field("labels", classes)
if anno and "segmentation" in anno[0]:
masks = [obj["segmentation"] for obj in anno]
masks = SegmentationMask(masks, img.size, mode='poly')
target.add_field("masks", masks)
if anno and "keypoints" in anno[0]:
keypoints = [obj["keypoints"] for obj in anno]
keypoints = PersonKeypoints(keypoints, img.size)
target.add_field("keypoints", keypoints)
target = target.clip_to_image(remove_empty=True)
if self._transforms is not None:
img, target = self._transforms(img, target)
return img, target, idx
def forward_for_single_feature_map(self, anchors, objectness, box_regression):
"""
Arguments:
anchors: list[BoxList]
objectness: tensor of size N, A, H, W
box_regression: tensor of size N, A * 4, H, W
"""
device = objectness.device
N, A, H, W = objectness.shape
# put in the same format as anchors
objectness = permute_and_flatten(objectness, N, A, 1, H, W).view(N, -1)
objectness = objectness.sigmoid()
box_regression = permute_and_flatten(box_regression, N, A, 4, H, W)
num_anchors = A * H * W
pre_nms_top_n = min(self.pre_nms_top_n, num_anchors)
objectness, topk_idx = objectness.topk(pre_nms_top_n, dim=1, sorted=True)
batch_idx = torch.arange(N, device=device)[:, None]
box_regression = box_regression[batch_idx, topk_idx]
image_shapes = [box.size for box in anchors]
concat_anchors = torch.cat([a.bbox for a in anchors], dim=0)
concat_anchors = concat_anchors.reshape(N, -1, 4)[batch_idx, topk_idx]
proposals = self.box_coder.decode(
box_regression.view(-1, 4), concat_anchors.view(-1, 4)
)
proposals = proposals.view(N, -1, 4)
result = []
for proposal, score, im_shape in zip(proposals, objectness, image_shapes):
boxlist = BoxList(proposal, im_shape, mode="xyxy")
boxlist.add_field("objectness", score)
boxlist = boxlist.clip_to_image(remove_empty=False)
boxlist = remove_small_boxes(boxlist, self.min_size)
boxlist = boxlist_nms(
boxlist,
self.nms_thresh,
max_proposals=self.post_nms_top_n,
score_field="objectness",
)
result.append(boxlist)
return result
def prepare_boxlist(self, boxes, scores, image_shape):
"""
Returns BoxList from `boxes` and adds probability scores information
as an extra field
`boxes` has shape (#detections, 4 * #classes), where each row represents
a list of predicted bounding boxes for each of the object classes in the
dataset (including the background class). The detections in each row
originate from the same object proposal.
`scores` has shape (#detection, #classes), where each row represents a list
of object detection confidence scores for each of the object classes in the
dataset (including the background class). `scores[i, j]`` corresponds to the
box at `boxes[i, j * 4:(j + 1) * 4]`.
"""
boxes = boxes.reshape(-1, 4)
scores = scores.reshape(-1)
boxlist = BoxList(boxes, image_shape, mode="xyxy")
boxlist.add_field("scores", scores)
return boxlist
def __getitem__(self, item):
img = Image.open(self.image_lists[item]).convert("RGB")
# dummy target
w, h = img.size
target = BoxList([[0, 0, w, h]], img.size, mode="xyxy")
if self.transforms is not None:
img, target = self.transforms(img, target)
return img, target
def forward(self, x, boxes):
"""
Arguments:
x (Tensor): the mask logits
boxes (list[BoxList]): bounding boxes that are used as
reference, one for each image
Returns:
results (list[BoxList]): one BoxList for each image, containing
the extra field mask
"""
mask_prob = x.sigmoid()
# select masks coresponding to the predicted classes
num_masks = x.shape[0]
labels = [bbox.get_field("labels") for bbox in boxes]
labels = torch.cat(labels)
index = torch.arange(num_masks, device=labels.device)
mask_prob = mask_prob[index, labels][:, None]
boxes_per_image = [len(box) for box in boxes]
mask_prob = mask_prob.split(boxes_per_image, dim=0)
if self.masker:
mask_prob = self.masker(mask_prob, boxes)
results = []
for prob, box in zip(mask_prob, boxes):
bbox = BoxList(box.bbox, box.size, mode="xyxy")
for field in box.fields():
bbox.add_field(field, box.get_field(field))
bbox.add_field("mask", prob)
results.append(bbox)
return results
def forward(self, image_list, feature_maps):
grid_sizes = [feature_map.shape[-2:] for feature_map in feature_maps]
anchors_over_all_feature_maps = self.grid_anchors(grid_sizes)
anchors = []
for i, (image_height,
image_width) in enumerate(image_list.image_sizes):
anchors_in_image = []
for anchors_per_feature_map in anchors_over_all_feature_maps:
boxlist = BoxList(anchors_per_feature_map,
(image_width, image_height),
mode="xyxy")
self.add_visibility_to(boxlist)
anchors_in_image.append(boxlist)
anchors.append(anchors_in_image)
return anchors
def get_groundtruth(self, index):
img_id = self.ids[index]
anno = ET.parse(self._annopath % img_id).getroot()
anno = self._preprocess_annotation(anno)
height, width = anno["im_info"]
target = BoxList(anno["boxes"], (width, height), mode="xyxy")
target.add_field("labels", anno["labels"])
target.add_field("difficult", anno["difficult"])
return target
def select_over_all_levels(self, boxlists):
num_images = len(boxlists)
results = []
for i in range(num_images):
scores = boxlists[i].get_field("scores")
labels = boxlists[i].get_field("labels")
boxes = boxlists[i].bbox
boxlist = boxlists[i]
result = []
# skip the background
for j in range(1, self.num_classes):
inds = (labels == j).nonzero().view(-1)
scores_j = scores[inds]
boxes_j = boxes[inds, :].view(-1, 4)
boxlist_for_class = BoxList(boxes_j, boxlist.size, mode="xyxy")
boxlist_for_class.add_field("scores", scores_j)
boxlist_for_class = boxlist_nms(
boxlist_for_class, self.nms_thresh,
score_field="scores"
)
num_labels = len(boxlist_for_class)
boxlist_for_class.add_field(
"labels", torch.full((num_labels,), j,
dtype=torch.int64,
device=scores.device)
)
result.append(boxlist_for_class)
result = cat_boxlist(result)
number_of_detections = len(result)
# Limit to max_per_image detections **over all classes**
if number_of_detections > self.fpn_post_nms_top_n > 0:
cls_scores = result.get_field("scores")
image_thresh, _ = torch.kthvalue(
cls_scores.cpu(),
number_of_detections - self.fpn_post_nms_top_n + 1
)
keep = cls_scores >= image_thresh.item()
keep = torch.nonzero(keep).squeeze(1)
result = result[keep]
results.append(result)
return results
def filter_results(self, boxlist, num_classes):
"""Returns bounding-box detection results by thresholding on scores and
applying non-maximum suppression (NMS).
"""
# unwrap the boxlist to avoid additional overhead.
# if we had multi-class NMS, we could perform this directly on the boxlist
boxes = boxlist.bbox.reshape(-1, num_classes * 4)
scores = boxlist.get_field("scores").reshape(-1, num_classes)
device = scores.device
result = []
# Apply threshold on detection probabilities and apply NMS
# Skip j = 0, because it's the background class
inds_all = scores > self.score_thresh
for j in range(1, num_classes):
inds = inds_all[:, j].nonzero().squeeze(1)
scores_j = scores[inds, j]
boxes_j = boxes[inds, j * 4:(j + 1) * 4]
boxlist_for_class = BoxList(boxes_j, boxlist.size, mode="xyxy")
boxlist_for_class.add_field("scores", scores_j)
boxlist_for_class = boxlist_nms(boxlist_for_class, self.nms)
num_labels = len(boxlist_for_class)
boxlist_for_class.add_field(
"labels",
torch.full((num_labels, ), j, dtype=torch.int64,
device=device))
result.append(boxlist_for_class)
result = cat_boxlist(result)
number_of_detections = len(result)
# Limit to max_per_image detections **over all classes**
if number_of_detections > self.detections_per_img > 0:
cls_scores = result.get_field("scores")
image_thresh, _ = torch.kthvalue(
cls_scores.cpu(),
number_of_detections - self.detections_per_img + 1)
keep = cls_scores >= image_thresh.item()
keep = torch.nonzero(keep).squeeze(1)
result = result[keep]
return result
def forward(self, x, boxes):
mask_prob = x
scores = None
if self.keypointer:
mask_prob, scores = self.keypointer(x, boxes)
assert len(boxes) == 1, "Only non-batched inference supported for now"
boxes_per_image = [box.bbox.size(0) for box in boxes]
mask_prob = mask_prob.split(boxes_per_image, dim=0)
scores = scores.split(boxes_per_image, dim=0)
results = []
for prob, box, score in zip(mask_prob, boxes, scores):
bbox = BoxList(box.bbox, box.size, mode="xyxy")
for field in box.fields():
bbox.add_field(field, box.get_field(field))
prob = PersonKeypoints(prob, box.size)
prob.add_field("logits", score)
bbox.add_field("keypoints", prob)
results.append(bbox)
return results
def __getitem__(self, idx):
img_path = self.img_paths[idx]
ann_path = self.ann_paths[idx]
if self.mode == "mask":
ann = torch.from_numpy(np.asarray(Image.open(ann_path)))
# masks are represented with tensors
boxes, segmentations, labels = self._processBinayMasks(ann)
else:
with open(ann_path, "r") as ann_file:
ann = json.load(ann_file)
# masks are represented with polygons
boxes, segmentations, labels = self._processPolygons(ann)
boxes, segmentations, labels = self._filterGT(boxes, segmentations, labels)
if len(segmentations) == 0:
empty_ann_path = self.get_img_info(idx)["ann_path"]
print("EMPTY ENTRY:", empty_ann_path)
# self.img_paths.pop(idx)
# self.ann_paths.pop(idx)
img, target, _ = self[(idx + 1) % len(self)]
# just override this image with the next
return img, target, idx
img = Image.open(img_path)
# Compose all into a BoxList instance
target = BoxList(boxes, img.size, mode="xyxy")
target.add_field("labels", torch.tensor(labels))
masks = SegmentationMask(segmentations, img.size, mode=self.mode)
target.add_field("masks", masks)
if self.transforms is not None:
img, target = self.transforms(img, target)
return img, target, idx
def evaluate_box_proposals(
predictions, dataset, 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 = []
num_pos = 0
for image_id, prediction in enumerate(predictions):
original_id = dataset.id_to_img_map[image_id]
img_info = dataset.get_img_info(image_id)
image_width = img_info["width"]
image_height = img_info["height"]
prediction = prediction.resize((image_width, image_height))
# sort predictions in descending order
# TODO maybe remove this and make it explicit in the documentation
inds = prediction.get_field("objectness").sort(descending=True)[1]
prediction = prediction[inds]
ann_ids = dataset.coco.getAnnIds(imgIds=original_id)
anno = dataset.coco.loadAnns(ann_ids)
gt_boxes = [obj["bbox"] for obj in anno if obj["iscrowd"] == 0]
gt_boxes = torch.as_tensor(gt_boxes).reshape(-1, 4) # guard against no boxes
gt_boxes = BoxList(gt_boxes, (image_width, image_height), mode="xywh").convert(
"xyxy"
)
gt_areas = torch.as_tensor([obj["area"] for obj in anno if obj["iscrowd"] == 0])
if len(gt_boxes) == 0:
continue
valid_gt_inds = (gt_areas >= area_range[0]) & (gt_areas <= area_range[1])
gt_boxes = gt_boxes[valid_gt_inds]
num_pos += len(gt_boxes)
if len(gt_boxes) == 0:
continue
if len(prediction) == 0:
continue
if limit is not None and len(prediction) > limit:
prediction = prediction[:limit]
overlaps = boxlist_iou(prediction, gt_boxes)
_gt_overlaps = torch.zeros(len(gt_boxes))
for j in range(min(len(prediction), len(gt_boxes))):
# find which proposal box maximally covers each gt box
# and get the iou amount of coverage for each gt box
max_overlaps, argmax_overlaps = overlaps.max(dim=0)
# find which gt box is 'best' covered (i.e. 'best' = most iou)
gt_ovr, gt_ind = max_overlaps.max(dim=0)
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.append(_gt_overlaps)
gt_overlaps = torch.cat(gt_overlaps, dim=0)
gt_overlaps, _ = torch.sort(gt_overlaps)
if thresholds is None:
step = 0.05
thresholds = torch.arange(0.5, 0.95 + 1e-5, step, dtype=torch.float32)
recalls = torch.zeros_like(thresholds)
# compute recall for each iou threshold
for i, t in enumerate(thresholds):
recalls[i] = (gt_overlaps >= t).float().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 forward_for_single_feature_map(
self, anchors, box_cls, box_regression):
"""
Arguments:
anchors: list[BoxList]
box_cls: tensor of size N, A * C, H, W
box_regression: tensor of size N, A * 4, H, W
"""
device = box_cls.device
N, _, H, W = box_cls.shape
A = box_regression.size(1) // 4
C = box_cls.size(1) // A
# put in the same format as anchors
box_cls = permute_and_flatten(box_cls, N, A, C, H, W)
box_cls = box_cls.sigmoid()
box_regression = permute_and_flatten(box_regression, N, A, 4, H, W)
num_anchors = A * H * W
candidate_inds = box_cls > self.pre_nms_thresh
pre_nms_top_n = candidate_inds.view(N, -1).sum(1)
pre_nms_top_n = pre_nms_top_n.clamp(max=self.pre_nms_top_n)
results = []
for per_box_cls, per_box_regression, per_pre_nms_top_n, \
per_candidate_inds, per_anchors in zip(
box_cls,
box_regression,
pre_nms_top_n,
candidate_inds,
anchors):
# Sort and select TopN
# TODO most of this can be made out of the loop for
# all images.
# TODO:Yang: Not easy to do. Because the numbers of detections are
# different in each image. Therefore, this part needs to be done
# per image.
per_box_cls = per_box_cls[per_candidate_inds]
per_box_cls, top_k_indices = \
per_box_cls.topk(per_pre_nms_top_n, sorted=False)
per_candidate_nonzeros = \
per_candidate_inds.nonzero()[top_k_indices, :]
per_box_loc = per_candidate_nonzeros[:, 0]
per_class = per_candidate_nonzeros[:, 1]
per_class += 1
detections = self.box_coder.decode(
per_box_regression[per_box_loc, :].view(-1, 4),
per_anchors.bbox[per_box_loc, :].view(-1, 4)
)
boxlist = BoxList(detections, per_anchors.size, mode="xyxy")
boxlist.add_field("labels", per_class)
boxlist.add_field("scores", per_box_cls)
boxlist = boxlist.clip_to_image(remove_empty=False)
boxlist = remove_small_boxes(boxlist, self.min_size)
results.append(boxlist)
return results
def calc_detection_voc_prec_rec(gt_boxlists, pred_boxlists, iou_thresh=0.5):
"""Calculate precision and recall based on evaluation code of PASCAL VOC.
This function calculates precision and recall of
predicted bounding boxes obtained from a dataset which has :math:`N`
images.
The code is based on the evaluation code used in PASCAL VOC Challenge.
"""
n_pos = defaultdict(int)
score = defaultdict(list)
match = defaultdict(list)
for gt_boxlist, pred_boxlist in zip(gt_boxlists, pred_boxlists):
pred_bbox = pred_boxlist.bbox.numpy()
pred_label = pred_boxlist.get_field("labels").numpy()
pred_score = pred_boxlist.get_field("scores").numpy()
gt_bbox = gt_boxlist.bbox.numpy()
gt_label = gt_boxlist.get_field("labels").numpy()
gt_difficult = gt_boxlist.get_field("difficult").numpy()
for l in np.unique(np.concatenate((pred_label, gt_label)).astype(int)):
pred_mask_l = pred_label == l
pred_bbox_l = pred_bbox[pred_mask_l]
pred_score_l = pred_score[pred_mask_l]
# sort by score
order = pred_score_l.argsort()[::-1]
pred_bbox_l = pred_bbox_l[order]
pred_score_l = pred_score_l[order]
gt_mask_l = gt_label == l
gt_bbox_l = gt_bbox[gt_mask_l]
gt_difficult_l = gt_difficult[gt_mask_l]
n_pos[l] += np.logical_not(gt_difficult_l).sum()
score[l].extend(pred_score_l)
if len(pred_bbox_l) == 0:
continue
if len(gt_bbox_l) == 0:
match[l].extend((0, ) * pred_bbox_l.shape[0])
continue
# VOC evaluation follows integer typed bounding boxes.
pred_bbox_l = pred_bbox_l.copy()
pred_bbox_l[:, 2:] += 1
gt_bbox_l = gt_bbox_l.copy()
gt_bbox_l[:, 2:] += 1
iou = boxlist_iou(
BoxList(pred_bbox_l, gt_boxlist.size),
BoxList(gt_bbox_l, gt_boxlist.size),
).numpy()
gt_index = iou.argmax(axis=1)
# set -1 if there is no matching ground truth
gt_index[iou.max(axis=1) < iou_thresh] = -1
del iou
selec = np.zeros(gt_bbox_l.shape[0], dtype=bool)
for gt_idx in gt_index:
if gt_idx >= 0:
if gt_difficult_l[gt_idx]:
match[l].append(-1)
else:
if not selec[gt_idx]:
match[l].append(1)
else:
match[l].append(0)
selec[gt_idx] = True
else:
match[l].append(0)
n_fg_class = max(n_pos.keys()) + 1
prec = [None] * n_fg_class
rec = [None] * n_fg_class
for l in n_pos.keys():
score_l = np.array(score[l])
match_l = np.array(match[l], dtype=np.int8)
order = score_l.argsort()[::-1]
match_l = match_l[order]
tp = np.cumsum(match_l == 1)
fp = np.cumsum(match_l == 0)
# If an element of fp + tp is 0,
# the corresponding element of prec[l] is nan.
prec[l] = tp / (fp + tp)
# If n_pos[l] is 0, rec[l] is None.
if n_pos[l] > 0:
rec[l] = tp / n_pos[l]
return prec, rec