def test_real_data(self):
self.depth_example, _ = load_pfm(
os.path.join('..', 'docs', 'img', 'pens.pfm'))
# test case with real data
plot_point_cloud(disp_arr=self.depth_example,
rgb_img=None,
down_scale=4,
show_axes=True)
plt.show()
def plot(self):
import matplotlib.pyplot as plt
fig, (ax1, ax2, ax3) = plt.subplots(1, 3)
ax1.set_title('disparity map')
ax1.imshow(self.disp_l, cmap='gray')
ax2.set_title('ground truth')
ax2.imshow(self.gt_l, cmap='gray')
ax3.set_title('difference')
ax3.imshow(self.disp_l - self.gt_l, cmap='gray')
plt.show()
plot_point_cloud(disp_arr=self.disp_l, rgb_img=self.img_l, down_scale=4)
plt.show()
def plot_point_cloud(self,
rgb_img=None,
down_scale: int = 4,
view_angles: (int, int) = (50, 70),
s: float = 0.5,
ax: Axes3D = None) -> Axes3D:
# use valid central view if rgb image is missing
rgb_img = self.central_view if rgb_img is None and self.central_view is not None else rgb_img
# downsample via interpolation
if down_scale >= 1:
rgb_img = img_resize(rgb_img.copy(), x_scale=1. /
down_scale) if rgb_img is not None else None
dpt_map = img_resize(self.depth_map.copy(),
x_scale=1. / down_scale)
else:
warnings.warn(
'Downscale parameter has to be a positive integer greater than zero.'
)
dpt_map = self.depth_map.copy()
if self.depth_map is not None:
ax = plot_point_cloud(dpt_map,
rgb_img=rgb_img,
down_scale=1,
view_angles=view_angles,
s=s,
ax=ax)
else:
self.sta.status_msg(msg='Depth map variable is empty')
ax = Axes3D(fig=plt.figure())
return ax
def test_point_cloud(self):
# test invalid downscale parameters
for i in range(-1, 1):
try:
plot_point_cloud(disp_arr=None, rgb_img=None, down_scale=i)
except IndexError as e:
self.assertTrue(e, IndexError)
# test case where rgb image missing
plot_point_cloud(disp_arr=np.ones([3, 3]), rgb_img=None, down_scale=1)
# test case for image dimension mismatch
plot_point_cloud(disp_arr=np.ones([6, 6]),
rgb_img=np.ones([3, 3, 3]),
down_scale=1)
# test valid case
plot_point_cloud(disp_arr=np.ones([6, 6]),
rgb_img=np.ones([6, 6, 3]),
down_scale=2)
# test invalid rgb values
plot_point_cloud(disp_arr=np.ones([6, 6]),
rgb_img=np.ones([6, 6, 3]) * -1,
down_scale=2)
# test Axes3D argument
fig, ax = plt.figure(), plt.axes(projection='3d')
ax_type = type(ax)
ax = plot_point_cloud(disp_arr=np.ones([6, 6]),
rgb_img=np.ones([6, 6, 3]),
view_angles=(50, 70),
ax=ax)
self.assertEqual(type(ax),
ax_type,
msg='Expected %s type, but got %s' %
(ax_type, type(ax)))
values = [float(v) / 1000 for v in list(range(5, 16, 1))]
fig, axs = plt.subplots(1, len(values), figsize=(20, 5))
for i in range(len(values)):
res, _ = primal_dual_algo(cisparity,
lambda_rof=0,
theta=3,
tau=values[i])
axs[i].imshow(res, cmap='gray')
axs[i].set_title(str(values[i]))
plt.show()
# plot results
fig, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(20, 5))
ax1.imshow(Normalizer(lf_img_arr[lf_c, lf_c, ...]).type_norm(),
cmap='gray')
ax2.imshow(Normalizer(cisparity).type_norm(), cmap='gray')
ax3.imshow(Normalizer(sisparity).type_norm(), cmap='gray')
plt.show()
plot_point_cloud(disp_arr=sisparity,
rgb_img=lf_img_arr[lf_c, lf_c, ...],
down_scale=4)
plt.show()
# save results
plt.imsave('depth.png', cisparity, cmap='gray')
save_pfm(sisparity, file_path='./depth.pfm', scale=1)
pts = disp2pts(sisparity, rgb_img=lf_img_arr[lf_c, lf_c, ...])
save_ply(pts, file_path='depth_1.ply')