def get_feature_map(self, im_patch, hog_cell_sz):
hog_feature = extract_hog_feature(im_patch,
cell_size=hog_cell_sz)[:, :, :27]
if hog_cell_sz > 1:
im_patch = self.mex_resize(
im_patch,
(self._window.shape[1], self._window.shape[0])).astype(
np.uint8)
gray = cv2.cvtColor(im_patch,
cv2.COLOR_BGR2GRAY)[:, :, np.newaxis] / 255 - 0.5
return np.concatenate((gray, hog_feature), axis=2)
def get_csr_features(img, center, scale, template_sz, resize_sz, cell_size):
center = (int(center[0]), int(center[1]))
patch = cv2.getRectSubPix(img,
patchSize=(int(scale * template_sz[0]),
int(scale * template_sz[1])),
center=center)
patch = cv2.resize(patch, resize_sz).astype(np.uint8)
hog_feature = extract_hog_feature(patch, cell_size)[:, :, :18]
# gray feature is included in the cn features
cn_feature = extract_cn_feature(patch, cell_size)
features = np.concatenate((hog_feature, cn_feature), axis=2)
return features
def get_scale_subwindow(self, im, center, base_target_sz, scale_factors, scale_window, scale_model_sz,
hog_scale_cell_sz):
n_scales = len(self.scale_factors)
out = None
for s in range(n_scales):
patch_sz = (int(base_target_sz[0] * scale_factors[s]),
int(base_target_sz[1] * scale_factors[s]))
patch_sz = (max(2, patch_sz[0]), max(2, patch_sz[1]))
im_patch = cv2.getRectSubPix(im, patch_sz, center)
im_patch_resized = self.mex_resize(im_patch, scale_model_sz).astype(np.uint8)
tmp = extract_hog_feature(im_patch_resized, cell_size=hog_scale_cell_sz)
if out is None:
out = tmp.flatten() * scale_window[s]
else:
out = np.c_[out, tmp.flatten() * scale_window[s]]
return out
def update(self, im, pos, base_target_sz, current_scale_factor):
patchL = cv2.getRectSubPix(
im, (int(np.floor(current_scale_factor * self.scale_sz[0])),
int(np.floor(current_scale_factor * self.scale_sz[1]))), pos)
patchL = cv2.resize(patchL, self.scale_sz_window)
# convert into logpolar
patchLp = cv2.logPolar(patchL.astype(np.float32),
((patchL.shape[1] - 1) / 2,
(patchL.shape[0] - 1) / 2),
self.mag,
flags=cv2.INTER_LINEAR + cv2.WARP_FILL_OUTLIERS)
patchLp = extract_hog_feature(patchLp, cell_size=4)
tmp_sc, _, _ = self.estimate_scale(self.model_patchLp, patchLp,
self.mag)
tmp_sc = np.clip(tmp_sc, a_min=0.6, a_max=1.4)
scale_factor = current_scale_factor * tmp_sc
self.model_patchLp = (
1 - self.learning_rate_scale
) * self.model_patchLp + self.learning_rate_scale * patchLp
return scale_factor
def _extract_scale_sample(self, im, pos, base_target_sz, scale_factors,
scale_model_sz):
scale_sample = []
base_target_sz = np.array([base_target_sz[0], base_target_sz[1]])
for idx, scale in enumerate(scale_factors):
patch_sz = np.floor(base_target_sz * scale)
im_patch = cv2.getRectSubPix(im,
(int(patch_sz[0]), int(patch_sz[1])),
pos)
if scale_model_sz[0] > patch_sz[1]:
interpolation = cv2.INTER_LINEAR
else:
interpolation = cv2.INTER_AREA
im_patch_resized = cv2.resize(
im_patch, (int(scale_model_sz[0]), int(scale_model_sz[1])),
interpolation=interpolation).astype(np.uint8)
scale_sample.append(
extract_hog_feature(im_patch_resized, cell_size=4).reshape(
(-1, 1)))
scale_sample = np.concatenate(scale_sample, axis=1)
return scale_sample
def init(self, im, pos, base_target_sz, current_scale_factor):
w, h = base_target_sz
avg_dim = (w + h) / 2.5
self.scale_sz = ((w + avg_dim) / current_scale_factor,
(h + avg_dim) / current_scale_factor)
self.scale_sz0 = self.scale_sz
self.cos_window_scale = cos_window(
(self.scale_sz_window[0], self.scale_sz_window[1]))
self.mag = self.cos_window_scale.shape[0] / np.log(
np.sqrt((self.cos_window_scale.shape[0]**2 +
self.cos_window_scale.shape[1]**2) / 4))
# scale lp
patchL = cv2.getRectSubPix(
im, (int(np.floor(current_scale_factor * self.scale_sz[0])),
int(np.floor(current_scale_factor * self.scale_sz[1]))), pos)
patchL = cv2.resize(patchL, self.scale_sz_window)
patchLp = cv2.logPolar(patchL.astype(np.float32),
((patchL.shape[1] - 1) / 2,
(patchL.shape[0] - 1) / 2),
self.mag,
flags=cv2.INTER_LINEAR + cv2.WARP_FILL_OUTLIERS)
self.model_patchLp = extract_hog_feature(patchLp, cell_size=4)
