def apply_nms(all_boxes,
thresh,
ignore_background=False,
boUsePythonImpl=True):
"""Apply non-maximum suppression to all predicted boxes output by the test_net method."""
num_classes = len(all_boxes)
num_images = len(all_boxes[0])
nms_boxes = [[[] for _ in range(num_images)] for _ in range(num_classes)]
nms_keepIndices = [[[] for _ in range(num_images)]
for _ in range(num_classes)]
for cls_ind in range(num_classes):
if ignore_background and (cls_ind == 0):
continue
for im_ind in range(num_images):
dets = all_boxes[cls_ind][im_ind]
if dets == []:
continue
if boUsePythonImpl:
keep = nmsPython(dets, thresh)
else:
keep = nms(dets, thresh)
if len(keep) == 0:
continue
nms_boxes[cls_ind][im_ind] = dets[keep, :].copy()
nms_keepIndices[cls_ind][im_ind] = keep
return nms_boxes, nms_keepIndices
def apply_nms(all_boxes, thresh, ignore_background=False, boUsePythonImpl=True):
"""Apply non-maximum suppression to all predicted boxes output by the test_net method."""
num_classes = len(all_boxes)
num_images = len(all_boxes[0])
nms_boxes = [[[] for _ in range(num_images)]
for _ in range(num_classes)]
nms_keepIndices = [[[] for _ in range(num_images)]
for _ in range(num_classes)]
for cls_ind in range(num_classes):
if ignore_background and (cls_ind == 0):
continue
for im_ind in range(num_images):
dets = all_boxes[cls_ind][im_ind]
if dets == []:
continue
if boUsePythonImpl:
keep = nmsPython(dets, thresh)
else:
keep = nms(dets, thresh)
if len(keep) == 0:
continue
nms_boxes[cls_ind][im_ind] = dets[keep, :].copy()
nms_keepIndices[cls_ind][im_ind] = keep
return nms_boxes, nms_keepIndices
def findSelectiveSearchRois(img, kvals, minSize, max_merging_iterations, nmsThreshold):
tmp = []
dlib.find_candidate_object_locations(imconvertCv2Ski(img), tmp, kvals, minSize, max_merging_iterations)
rois = [[d.left(), d.top(), d.right(), d.bottom()] for d in tmp]
if nmsThreshold != None:
assert(nmsThreshold > 0 and nmsThreshold < 1)
dets = [ToFloats(r) + [abs((r[2] - r[0]) * (r[3] - r[1]))] for r in rois]
keepInds = nmsPython(np.array(dets), nmsThreshold)
#print("findSelectiveSearchRois using nms threshold: {}: before nms nrRois={}, after nms nrRois={}".format(nmsThreshold, len(rois), len(keepInds)))
#groupedRectangles, weights = cv2.groupRectangles(np.asanyarray(rectsInput, np.float).tolist(), 1, 0.3)
rois = [rois[i] for i in keepInds]
random.shuffle(rois) # randomize ROI order to not introduce any unintended effects later
return rois
