def op_script(data: torch.Tensor) -> torch.Tensor:
return kornia.rgb_to_hsv(data)
data = torch.tensor([[[[21., 22.], [22., 22.]],
[[13., 14.], [14., 14.]],
[[8., 8.], [8., 8.]]]]) # 3x2x2
actual = op_script(data)
expected = kornia.rgb_to_hsv(data)
assert_allclose(actual, expected)
def augment(im, aug_type):
''' Augment images as per given augmentation type. '''
if aug_type == 'normal':
im = im
elif aug_type == 'rotated':
rot_angle = np.random.choice([-90, 90, 180])
im = kornia_affine(im, rot_angle, 'rotate')
elif aug_type == 'hsv':
adjustFactor = np.random.choice(
[-0.1, -0.075, -0.05, 0.05, 0.075, 0.1])
hue, sat, value = torch.chunk(kornia.rgb_to_hsv(im), chunks=3, dim=-3)
adjust_mat = (torch.ones(1, 512, 512) * adjustFactor).cuda()
hueNew = hue + hue * adjust_mat.cuda()
hueNew = torch.clamp(hueNew, -2 * np.pi, 2 * np.pi)
satNew = sat + sat * adjust_mat.cuda()
satNew = torch.clamp(satNew, -2 * np.pi, 2 * np.pi)
new_im = torch.cat([hueNew, satNew, value], dim=-3)
im = kornia.hsv_to_rgb(new_im)
elif aug_type == 'gauss_noise':
im = augment_gaussian_noise(im)
elif aug_type == 'mirror':
im = torch.flip(im, [-2])
return im
def __call__(self,x):
x = kornia.rgb_to_hsv(x)
return x
#############################
# Create a batch of images
xb_bgr = torch.cat([x_bgr, hflip(x_bgr), vflip(x_bgr), rot180(x_bgr)])
imshow(xb_bgr)
#############################
# Convert BGR to RGB
xb_rgb = kornia.bgr_to_rgb(xb_bgr)
imshow(xb_rgb)
#############################
# Convert RGB to grayscale
# NOTE: image comes in torch.uint8, and kornia assumes floating point type
xb_gray = kornia.rgb_to_grayscale(xb_rgb.float() / 255.0)
imshow(xb_gray)
#############################
# Convert RGB to HSV
xb_hsv = kornia.rgb_to_hsv(xb_rgb.float() / 255.0)
imshow(xb_hsv[:, 2:3])
#############################
# Convert RGB to YUV
# NOTE: image comes in torch.uint8, and kornia assumes floating point type
yuv = kornia.rgb_to_yuv(xb_rgb.float() / 255.0)
y_channel = torchvision.utils.make_grid(yuv, nrow=2)[0, :, :]
plt.imshow(y_channel, cmap='gray', vmin=0, vmax=1) # Displaying only y channel
plt.axis('off')
plt.show()
def _rgb_to_hsv(img: torch.Tensor) -> torch.Tensor:
return kornia.rgb_to_hsv(img)