def __init__(self, tracker_name, imagefile, region):
# param
self.hp, self.evaluation, self.run, self.env, self.design = parse_arguments(
)
self.final_score_sz = 273
# init network
self.siam = SiameseNet(tracker_name)
if tracker_name == 'SiamFC':
pretrained = None
elif tracker_name == 'SiamVGG':
pretrained = None
elif tracker_name == 'SiamFCRes22':
pretrained = None
elif tracker_name == 'SiamFCIncep22':
pretrained = None
elif tracker_name == 'SiamFCNext22':
pretrained = './cp/siamresnextcheckpoint.pth.tar'
load_net(pretrained, self.siam)
self.siam.cuda()
# init bbox
bbox = convert_bbox_format(region, 'center-based')
self.pos_x, self.pos_y, self.target_w, self.target_h = bbox.x, bbox.y, bbox.width, bbox.height
# init scale factor, penalty
self.scale_factors = self.hp.scale_step**np.linspace(
-np.ceil(self.hp.scale_num / 2), np.ceil(self.hp.scale_num / 2),
self.hp.scale_num)
hann_1d = np.expand_dims(np.hanning(self.final_score_sz), axis=0)
self.penalty = np.transpose(hann_1d) * hann_1d
self.penalty = self.penalty / np.sum(self.penalty)
context = self.design.context * (self.target_w + self.target_h)
self.z_sz = np.sqrt(
np.prod((self.target_w + context) * (self.target_h + context)))
self.x_sz = float(
self.design.search_sz) / self.design.exemplar_sz * self.z_sz
image_, self.templates_z_ = self.siam.get_template_z_new(
self.pos_x, self.pos_y, self.z_sz, imagefile, self.design)
class SiamVGGTracker(object):
def __init__(self, tracker_name, imagefile, region):
# param
self.hp, self.evaluation, self.run, self.env, self.design = parse_arguments(
)
self.final_score_sz = 273
# init network
self.siam = SiameseNet(tracker_name)
if tracker_name == 'SiamFC':
pretrained = None
elif tracker_name == 'SiamVGG':
pretrained = None
elif tracker_name == 'SiamFCRes22':
pretrained = None
elif tracker_name == 'SiamFCIncep22':
pretrained = None
elif tracker_name == 'SiamFCNext22':
pretrained = './cp/siamresnextcheckpoint.pth.tar'
load_net(pretrained, self.siam)
self.siam.cuda()
# init bbox
bbox = convert_bbox_format(region, 'center-based')
self.pos_x, self.pos_y, self.target_w, self.target_h = bbox.x, bbox.y, bbox.width, bbox.height
# init scale factor, penalty
self.scale_factors = self.hp.scale_step**np.linspace(
-np.ceil(self.hp.scale_num / 2), np.ceil(self.hp.scale_num / 2),
self.hp.scale_num)
hann_1d = np.expand_dims(np.hanning(self.final_score_sz), axis=0)
self.penalty = np.transpose(hann_1d) * hann_1d
self.penalty = self.penalty / np.sum(self.penalty)
context = self.design.context * (self.target_w + self.target_h)
self.z_sz = np.sqrt(
np.prod((self.target_w + context) * (self.target_h + context)))
self.x_sz = float(
self.design.search_sz) / self.design.exemplar_sz * self.z_sz
image_, self.templates_z_ = self.siam.get_template_z_new(
self.pos_x, self.pos_y, self.z_sz, imagefile, self.design)
def track(self, imagefile):
scaled_search_area = self.x_sz * self.scale_factors
scaled_target_w = self.target_w * self.scale_factors
scaled_target_h = self.target_h * self.scale_factors
image_, scores_ = self.siam.get_scores_new(self.pos_x, self.pos_y,
scaled_search_area,
self.templates_z_,
imagefile, self.design,
self.final_score_sz)
scores_ = np.squeeze(scores_)
scores_[0, :, :] = self.hp.scale_penalty * scores_[0, :, :]
scores_[2, :, :] = self.hp.scale_penalty * scores_[2, :, :]
new_scale_id = np.argmax(np.amax(scores_, axis=(1, 2)))
self.x_sz = (
1 - self.hp.scale_lr
) * self.x_sz + self.hp.scale_lr * scaled_search_area[new_scale_id]
self.target_w = (
1 - self.hp.scale_lr
) * self.target_w + self.hp.scale_lr * scaled_target_w[new_scale_id]
self.target_h = (
1 - self.hp.scale_lr
) * self.target_h + self.hp.scale_lr * scaled_target_h[new_scale_id]
score_ = scores_[new_scale_id, :, :]
score_ = score_ - np.min(score_)
score_ = score_ / np.sum(score_)
score_ = (1 - self.hp.window_influence
) * score_ + self.hp.window_influence * self.penalty
self.pos_x, self.pos_y = _update_target_position(
self.pos_x, self.pos_y, score_, self.final_score_sz,
self.design.tot_stride, self.design.search_sz, self.hp.response_up,
self.x_sz)
bbox = Rectangle(self.pos_x, self.pos_y, self.target_w, self.target_h)
bbox = convert_bbox_format(bbox, 'top-left-based')
return bbox, np.max(score_)