3jSSKJr SSKJr SSKJrJrJr SSKrSSK J s J r SSKJ r SSK JrJrJr SSKJrJrJr SSKJr SS KJrJr S6S jrS6S jrS6S jrS6S jrS6SjrS7S8SjjrS9S:Sjjr S6Sjr!S9S:Sjjr"S9S:Sjjr#S;SSjjr%S?S@Sjjr&SASBSjjr'\SCSj5r(\SDSj5r)SESjr*SFSjr+SGSjr,\SHSj5r-\SHSj5r.SISJSjjr/"S S!\ R`5r1"S"S#\ R`5r2"S$S%\ R`5r3"S&S'\ R`5r4"S(S)\ R`5r5"S*S+\ R`5r6"S,S-\ R`5r7"S.S/\ R`5r8"S0S1\ R`5r9"S2S3\ R`5r:"S4S5\ R`5r;g)K) annotations)pi)ClassVarOptionalUnionN)nn) hsv_to_rgbrgb_to_grayscale rgb_to_hsv) KORNIA_CHECKKORNIA_CHECK_IS_COLOR_OR_GRAYKORNIA_CHECK_IS_TENSOR)_torch_histc_cast)perform_keep_shape_imageperform_keep_shape_videoc[US5 [[U[[R 45S5 [U[5(a*[R "XRURS9nOE[U[R 5(a&URURUR5n[UR5[UR5:wa3USn[UR5[UR5:waM3[R"USSS9up#n[R"X1-SS S 9n[R"X%U/SS 9nU$) zSAdjust color saturation of an image. Expecting image to be in hsv format already. Expected shape (*, H, W)'Factor should be float or torch.Tensor.devicedtype.Nchunksdimrminmaxr)rr isinstancefloattorchTensor as_tensorrrtolenshapechunkclampcat)imagefactorhsvs_outouts O/home/wildlama/miniconda3/lib/python3.13/site-packages/kornia/enhance/adjust.pyadjust_saturation_rawr6%s 5"<=FUELL$9:>> x = torch.ones(1, 3, 3, 3) >>> adjust_saturation_with_gray_subtraction(x, 2.).shape torch.Size([1, 3, 3, 3]) >>> x = torch.ones(2, 3, 3, 3) >>> y = torch.tensor([1., 2.]) >>> adjust_saturation_with_gray_subtraction(x, y).shape torch.Size([2, 3, 3, 3]) rrz$Image should be an RGB or gray imagerrrr?)rr r#r$r%r&r r*r'rrr(r)r r,)r.r/x_other x_adjustedr4s r5'adjust_saturation_with_gray_subtractionr=Cs:5"<=FUELL$9:`__. Example: >>> x = torch.ones(1, 3, 3, 3) >>> adjust_saturation(x, 2.).shape torch.Size([1, 3, 3, 3]) >>> x = torch.ones(2, 3, 3, 3) >>> y = torch.tensor([1., 2.]) >>> adjust_saturation(x, y).shape torch.Size([2, 3, 3, 3]) )r r6r r.r/x_hsvr<r4s r5adjust_saturationrA|s+>%U+E 5UCJ#:.C Jr7c[US5 [[U[[R 45S[ U535 [U[5(a[R"U5nURURUR5n[UR5[UR5:wa3USn[UR5[UR5:waM3[R"USSS9up#nS[-n[R"X!-U5n[R "XcU/SS9nU$) zFAdjust hue of an image. Expecting image to be in hsv format already. rzXThe factor should be a float number or torch.Tensor in the range between [-PI, PI]. Got rrrrr")rr r#r$r%r&typer'r(rrr)r*r+rfmodr-)r.r/r0r1r2divisorh_outr4s r5adjust_hue_rawrHs 5"<=6E5<<01 bcghncobpq &%  ( YYu||U[[ 1F fll s5;;/ / " fll s5;;/ /kk%r2GA!VG**QZ9E 5Q-R8C Jr7cH[U5n[X!5n[U5nU$)aAdjust hue of an image. .. image:: _static/img/adjust_hue.png The image is expected to be an RGB image in the range of [0, 1]. Args: image: Image to be adjusted in the shape of :math:`(*, 3, H, W)`. factor: How much to shift the hue channel. Should be in [-PI, PI]. PI and -PI give complete reversal of hue channel in HSV space in positive and negative direction respectively. 0 means no shift. Therefore, both -PI and PI will give an image with complementary colors while 0 gives the original image. Return: Adjusted image in the shape of :math:`(*, 3, H, W)`. .. note:: See a working example `here `__. Example: >>> x = torch.ones(1, 3, 2, 2) >>> adjust_hue(x, 3.141516).shape torch.Size([1, 3, 2, 2]) >>> x = torch.ones(2, 3, 3, 3) >>> y = torch.ones(2) * 3.141516 >>> adjust_hue(x, y).shape torch.Size([2, 3, 3, 3]) )r rHr r?s r5 adjust_huerJs,@%U+E .e`__. Example: >>> x = torch.ones(1, 1, 2, 2) >>> adjust_gamma(x, 1.0, 2.0) tensor([[[[1., 1.], [1., 1.]]]]) >>> x = torch.ones(2, 5, 3, 3) >>> y1 = torch.ones(2) * 1.0 >>> y2 = torch.ones(2) * 2.0 >>> adjust_gamma(x, y1, y2).shape torch.Size([2, 5, 3, 3]) &Input type is not a torch.Tensor. Got z:The gamma should be a positive float or torch.Tensor. Got z9The gain should be a positive float or torch.Tensor. Got r9z Gamma must be non-negative. Got zGain must be non-negative. Got r"r:)r#r%r& TypeErrorrDr$r(rrany ValueErrorranger)r* unsqueezepowr,)inputgammagain_x_adjustr4s r5 adjust_gammarYsF eU\\ * *@e NOO eeU\\2 3 3TUYZ_U`Tabcc dUELL1 2 2STXY]T^S_`aa% eW%$||TF# HHU\\ " % %ekk 2E 775<< # #EKK 0D  ;E7CDD s :4&ABB 3u{{#c%++&66 72.83u{{#c$**o5 6t,7"EIIe$;;H Hc37C Jr7c[US5 [[U[[R 45S5 [U[5(a*[R "XRURS9nOE[U[R 5(a&URURUR5n[UR5[UR5:wa3USn[UR5[UR5:waM3[[US:5S5 X-nU(aURSSS 9nU$) a@Adjust the contrast of an image torch.Tensor. .. image:: _static/img/adjust_contrast.png This implementation follows Szeliski's book convention, where contrast is defined as a `multiplicative` operation directly to raw pixel values. Beware that other frameworks might use different conventions which can be difficult to reproduce exact results. The input image and factor is expected to be in the range of [0, 1]. .. tip:: This is not the preferred way to adjust the contrast of an image. Ideally one must implement :func:`kornia.enhance.adjust_gamma`. More details in the following link: https://scikit-image.org/docs/dev/auto_examples/color_exposure/plot_log_gamma.html#sphx-glr-auto-examples-color-exposure-plot-log-gamma-py Args: image: Image to be adjusted in the shape of :math:`(*, H, W)`. factor: Contrast adjust factor per element in the batch. 0 generates a completely black image, 1 does not modify the input image while any other non-negative number modify the brightness by this factor. clip_output: whether to clip the output image with range of [0, 1]. Return: Adjusted image in the shape of :math:`(*, H, W)`. .. note:: See a working example `here `__. Example: >>> import torch >>> x = torch.ones(1, 1, 2, 2) >>> adjust_contrast(x, 0.5) tensor([[[[0.5000, 0.5000], [0.5000, 0.5000]]]]) >>> x = torch.ones(2, 5, 3, 3) >>> y = torch.tensor([0.65, 0.50]) >>> adjust_contrast(x, y).shape torch.Size([2, 5, 3, 3]) rrrrrz!Contrast factor must be positive.r9r:r)rr r#r$r%r&r'rrr(r)r*rOr,r.r/ clip_output img_adjusts r5adjust_contrastr^<sV5"<=FUELL$9:>> import torch >>> x = torch.ones(1, 1, 2, 2) >>> adjust_contrast_with_mean_subtraction(x, 0.5) tensor([[[[1., 1.], [1., 1.]]]]) >>> x = torch.ones(2, 5, 3, 3) >>> y = torch.tensor([0.65, 0.50]) >>> adjust_contrast_with_mean_subtraction(x, y).shape torch.Size([2, 5, 3, 3]) rrrrrr)rMTrr9r:r)rr r#r$r%r&r'rrr(r)r*r meanr,)r.r/img_meanr]s r5%adjust_contrast_with_mean_subtractionrcs <5"<=FUELL$9:`__. Example: >>> x = torch.ones(1, 1, 2, 2) >>> adjust_brightness(x, 1.) tensor([[[[1., 1.], [1., 1.]]]]) >>> x = torch.ones(2, 5, 3, 3) >>> y = torch.tensor([0.25, 0.50]) >>> adjust_brightness(x, y).shape torch.Size([2, 5, 3, 3]) rrrrr9r:r rr r#r$r%r&r'rrr(r)r*r,r[s r5adjust_brightnessrfs\5"<=FUELL$9:>> x = torch.ones(1, 1, 2, 2) >>> adjust_brightness_accumulative(x, 1.) tensor([[[[1., 1.], [1., 1.]]]]) >>> x = torch.ones(2, 5, 3, 3) >>> y = torch.tensor([0.25, 0.50]) >>> adjust_brightness_accumulative(x, y).shape torch.Size([2, 5, 3, 3]) rrrrr9r:rrer[s r5adjust_brightness_accumulativerhs>5"<=FUELL$9:>> x = torch.ones(1, 1, 2, 2) >>> adjust_sigmoid(x, gain=0) tensor([[[[0.5000, 0.5000], [0.5000, 0.5000]]]]) rr)rexp)r.cutoffrVinvr]s r5adjust_sigmoidrm7se25"<= a4E>#:"?"?"AABB  !t~6;;==> r7c[US5 U(a SU-S- U-nOSU-R5U-nU(aURSSS9nU$)aAdjust log correction on the input image torch.Tensor. The input image is expected to be in the range of [0, 1]. Reference: [1]: http://www.ece.ucsb.edu/Faculty/Manjunath/courses/ece178W03/EnhancePart1.pdf Args: image: Image to be adjusted in the shape of :math:`(*, H, W)`. gain: The multiplier of logarithmic function. inv: If is set to True the function will return the inverse logarithmic correction. clip_output: Whether to clip the output image with range of [0, 1]. Returns: Adjusted torch.Tensor in the shape of :math:`(*, H, W)`. Example: >>> x = torch.zeros(1, 1, 2, 2) >>> adjust_log(x, inv=True) tensor([[[[0., 0.], [0., 0.]]]]) rrCrr9r:r)rlog2r,)r.rVrlr\r]s r5 adjust_logrpYs[05"<= hld* %i%%'$. %%#3%7 r7c 0[U[R5(d[S[ U535e[U[ [R45(d[S[ U535e[U[R5(a[ UR5S:waURS5[ U5:Xa[ UR5S:Xd [SURS5SU35eURUR5RUR5n[R"UVs/sHo"R"URSS6PM sn5n[R"X:US U- 5$s snf) aFor each pixel in the image, select the pixel if the value is less than the threshold. Otherwise, subtract 1.0 from the pixel. Args: input: image or batched images to solarize. thresholds: solarize thresholds. If int or one element torch.Tensor, input will be solarized across the whole batch. If 1-d torch.Tensor, input will be solarized element-wise, len(thresholds) == len(input). Returns: Solarized images. rL9The factor should be either a float or torch.Tensor. Got rrz*thresholds must be a 1-d vector of shape (,). Got rNr:)r#r%r&rNrDr$r)r*sizeAssertionErrorr(rrstackexpandwhere)rT thresholdsxs r5 _solarizer{s; eU\\ * *@e NOO j5%,,"7 8 8STXYcTdSefgg*ell++J4D4D0E0J 1 Z0S9I9I5Ja5O #MejjYZm_\deodp!qr r]]5<<033EKK@ [[z!Rz!((EKK,<"=z!RS ;;u)5#+ >>"Ss %Fc [U[R5(d[S[ U535e[U[ [R45(d[S[ U535e[U[ 5(a[R "U5nUGb[U[ [R45(d[S[ U535e[U[ 5(a[R "U5n[R"US:US:-5(d[SUS35e[U[R5(a[UR5S:waURS5[U5:Xa[UR5S :Xd [S URS5S U35eURUR5RUR5n[R"UVs/sHo3R "URS S6PM sn5nX-nUR#S S5n[%X5$s snf)aFor each pixel in the image less than threshold. .. image:: _static/img/solarize.png We add 'addition' amount to it and then clip the pixel value to be between 0 and 1.0. The value of 'addition' is between -0.5 and 0.5. Args: input: image torch.Tensor with shapes like :math:`(*, C, H, W)` to solarize. thresholds: solarize thresholds. If int or one element torch.Tensor, input will be solarized across the whole batch. If 1-d torch.Tensor, input will be solarized element-wise, len(thresholds) == len(input). additions: between -0.5 and 0.5. If None, no addition will be performed. If int or one element torch.Tensor, same addition will be added across the whole batch. If 1-d torch.Tensor, additions will be added element-wisely, len(additions) == len(input). Returns: The solarized images with shape :math:`(*, C, H, W)`. Example: >>> x = torch.rand(1, 4, 3, 3) >>> out = solarize(x, thresholds=0.5, additions=0.) >>> out.shape torch.Size([1, 4, 3, 3]) >>> x = torch.rand(2, 4, 3, 3) >>> thresholds = torch.tensor([0.8, 0.5]) >>> additions = torch.tensor([-0.25, 0.25]) >>> solarize(x, thresholds, additions).shape torch.Size([2, 4, 3, 3]) rLrrN?gz5The value of 'addition' is between -0.5 and 0.5. Got .rrz)additions must be a 1-d vector of shape (rsrr9r:)r#r%r&rNrDr$r'allrur)r*rtr(rrrvrwr,r{)rTry additionsrzs r5solarizersL eU\\ * *@e NOO j5%,,"7 8 8STXYcTdSefgg*e$$__Z0 )eU\\%:;;WX\]fXgWhij j i ' ' 2Iyy)c/i$.>?@@ #XYbXccd!ef f i . .3y3G13LJJqMS^3IOO8LPQ8Q$'PQVQ[Q[\]Q^P__ghqgr%stt! U\\255ekkBI )$T)QXXu{{23/?%@)$TUI! C% U '' %Us%I c ^^[U[R5(d[S[ U535e[U[ [R45(d[S[ U535e[U[ 5(a[R "U5nSSjmSSjmSUU4Sjjn[UR5S:Xd([UR5S:Xa[U5S:XaU"X5$/n[UR5S:XaURSURS:wa,[SURSS URSS 35e[URS5HnURU"XX55 M! [R"USS 9$URURS [UR5:wa<[S URS URS [UR5S 35eUR"S/UR[UR5S Q76nUR5n[URS5HnURU"XTXd55 M! [R"USS 9R"UR6$)aReduce the number of bits for each color channel. .. image:: _static/img/posterize.png Non-differentiable function, ``torch.uint8`` involved. Args: input: image torch.Tensor with shape :math:`(*, C, H, W)` to posterize. bits: number of high bits. Must be in range [0, 8]. If int or one element torch.Tensor, input will be posterized by this bits. If 1-d torch.Tensor, input will be posterized element-wisely, len(bits) == input.shape[-3]. If n-d torch.Tensor, input will be posterized element-channel-wisely, bits.shape == input.shape[:len(bits.shape)] Returns: Image with reduced color channels with shape :math:`(*, C, H, W)`. Example: >>> x = torch.rand(1, 6, 3, 3) >>> out = posterize(x, bits=8) >>> torch.testing.assert_close(x, out) >>> x = torch.rand(2, 6, 3, 3) >>> bits = torch.tensor([4, 2]) >>> posterize(x, bits).shape torch.Size([2, 6, 3, 3]) rLz-bits type is not an int or torch.Tensor. Got cUS-R[R5SU--RUR5S- $NrCo@r(r%uint8rrTshifts r5 _left_shiftposterize.._left_shifts8  -E:>>u{{KeSSr7cUS-R[R5SU-RUR5- S- $rrrs r5 _right_shiftposterize.._right_shifts6  ,5}}U[[/IIEQQr7c>US:Xa[R"U5$US:XaUR5$SU- nT"T"X5U5$)Nr)r% zeros_likeclone)rTbitsrrs r5_posterize_one!posterize.._posterize_onesI 19##E* * 19;;= 4x<4d;;r7rrz5Batch size must be equal between bits and input. Got z, r~r"Nzs4::!? @ C  ZZ " >&)UX67# ;;s " * *EKK 88r7c [U[R5(d)[R"XRUR S9n[ UR55S:wa^UR[R"URS5/5:wa*[SURS5SUR35e[R"/SQ/SQ/SQ/UR URS9RSSS S 5RURS5SSS5S - n[RRRXS SURS5S 9n[R "US S5n[R""U5n[$R&"U/SQ5n[$R&"U/SQ5n[R("US:HX`5n[ UR55S:Xa [+XpU5$[R,"[/[ U55Vs/sHn[+XxXX5PM sn5$s snf)aApply sharpness to the input torch.Tensor. .. image:: _static/img/sharpness.png Implemented Sharpness function from PIL using torch ops. This implementation refers to: https://github.com/tensorflow/tpu/blob/master/models/official/efficientnet/autoaugment.py#L326 Args: input: image torch.Tensor with shape :math:`(*, C, H, W)` to sharpen. factor: factor of sharpness strength. Must be above 0. If float or one element torch.Tensor, input will be sharpened by the same factor across the whole batch. If 1-d torch.Tensor, input will be sharpened element-wisely, len(factor) == len(input). Returns: Sharpened image or images with shape :math:`(*, C, H, W)`. Example: >>> x = torch.rand(1, 1, 5, 5) >>> sharpness(x, 0.5).shape torch.Size([1, 1, 5, 5]) rrzYInput batch size shall match with factor size if factor is not a 0-dim torch.Tensor. Got z and )rrr)rr)rrrr N)biasstridegroupsr9r:)rrrr)r#r%r&r'rrr)rtr*Sizerurrepeatr functionalconv2dr, ones_likeFpadrx _blend_onervrQ) rTr/kernel degeneratemask padded_maskpadded_degenerateresultrs r5 sharpnessr=s0 fell + + EKKP 6;;=Q6<<5::uzz!}o3N#N ::a=/v||n 6  Iy9UZUaUab aAq   1 q!Q '   $$++EQW\WaWabcWd+eJZc2J ??: &D%%l+Kj,7 [[)+< DF 6;;=Q&00 ;;ERUV\R]L^_L^q 69eh BL^_ ``_s/Ic[U[R5(d[SUS35e[U[R5(d[SUS35e[U[R5(a,[ UR 55S:wa[SUS35eUS:XaU$US:XaU$X- U-nX-nUS:aUS:aU$[R "USS5$)a<Blend two images into one. Args: input1: image torch.Tensor with shapes like :math:`(H, W)` or :math:`(D, H, W)`. input2: image torch.Tensor with shapes like :math:`(H, W)` or :math:`(D, H, W)`. factor: factor 0-dim torch.Tensor. Returns: : image torch.Tensor with the batch in the zero position. z%`input1` must be a torch.Tensor. Got r~rz!,C[\b[ccdeff }  } Ov %D -C |  ;;sAq !!r7c@[R"USSS9n[R"[R"US5U-USS9n[R"[R"SUR UR S9USS/5n[R"USS5$) NrCtrunc rounding_moderrrrMr)r%divcumsumr-zerosrrr,)histostep step_truncluts r5 _build_lutrs{4':J ))ELL*Z7W UC ))U[[3::SYYGSbR SC ;;sAs ##r7cUR5nUR5nUR5S:aV[R"U[R "SUR S95(d[SUR5S35eUR5S:aV[R"U[R "SUR S95(d[SUR5S35e[UR5nUS;a[SUS35eUS -n[US S S S 9n[R"XDS :gS/5n[R"[R"U5US- S SS9nUS :XaUnUS - $[R"[!XF5S UR#5R%55nUR'U5nUS - $)zScale the data in the channel to implement equalize. Args: im: image torch.Tensor with shapes like :math:`(H, W)` or :math:`(D, H, W)`. Returns: image torch.Tensor with the batch in the zero position. r9)rzEValues in the input torch.Tensor must greater or equal to 0.0. Found r~r:zCValues in the input torch.Tensor must lower or equal to 1.0. Found )rCrz6Input torch.Tensor must have 2 or 3 dimensions. Found rrr)binsr r!rMrr)r r!itemr%iscloser'rrPr)r*rNrrrsumgatherrrlong reshape_as)immin_max_ndimsr nonzero_historrs r5_scale_channelrs 668D 668D yy{StU__SPTPZPZ5[!\!\`aeajajal`mmnopp yy{StU__SPTPZPZ5[!\!\^_c_h_h_j^kklmnn ME FPQVPWWXYZZ eB bss ;EMM% "3bT:M 99UYY}- b0AA3V] ^D qy E>j5q"**,:K:K:MN""2& E>r7c /nUH_n[R"[[U55Vs/sHn[ X#SS2SS245PM sn5nUR U5 Ma [R"U5$s snf)a#Apply equalize on the input torch.Tensor. .. image:: _static/img/equalize.png Implements Equalize function from PIL using PyTorch ops based on uint8 format: https://github.com/tensorflow/tpu/blob/5f71c12a020403f863434e96982a840578fdd127/models/official/efficientnet/autoaugment.py#L355 Args: input: image torch.Tensor to equalize with shape :math:`(*, C, H, W)`. Returns: Equalized image torch.Tensor with shape :math:`(*, C, H, W)`. Example: >>> x = torch.rand(1, 2, 3, 3) >>> equalize(x).shape torch.Size([1, 2, 3, 3]) Nr%rvrQr)rr)rTrr.r scaled_images r5equalizersp* C{{ERUV[R\L]#^L]qN5Aq>$BL]#^_  <  ;;s $_sA> c /nUHbn[R"[[U55Vs/sHn[ X#SS2SS2SS245PM sn5nUR U5 Md [R"U5$s snf)aEqualize the values for a 3D volumetric torch.Tensor. Implements Equalize function for a sequence of images using PyTorch ops based on uint8 format: https://github.com/tensorflow/tpu/blob/master/models/official/efficientnet/autoaugment.py#L352 Args: input: image torch.Tensor with shape :math:`(*, C, D, H, W)` to equalize. Returns: Equalized volume with shape :math:`(B, C, D, H, W)`. Nr)rTrvolumer scaled_inputs r5 equalize3drst C{{PUVYZ`VaPb#cPb1N6Q1*3E$FPb#cd  <  ;;s $ds!B cD[U[R5(d[S[ [ 535eUc[R "S/5nOUn[U[R5(d[S[ U535eURU5U- $)aInvert the values of an input image torch.Tensor by its maximum value. .. image:: _static/img/invert.png Args: image: The input torch.Tensor to invert with an arbitatry shape. max_val: The expected maximum value in the input torch.Tensor. The shape has to according to the input torch.Tensor shape, or at least has to work with broadcasting. Example: >>> img = torch.rand(1, 2, 4, 4) >>> invert(img).shape torch.Size([1, 2, 4, 4]) >>> img = 255. * torch.rand(1, 2, 3, 4, 4) >>> invert(img, torch.as_tensor(255.)).shape torch.Size([1, 2, 3, 4, 4]) >>> img = torch.rand(1, 3, 4, 4) >>> invert(img, torch.as_tensor([[[[1.]]]])).shape torch.Size([1, 3, 4, 4]) z"Input is not a torch.Tensor. Got: r:z$max_val is not a torch.Tensor. Got: )r#r%r&rurDrTtensorr()r.max_val_max_vals r5invertrs0 eU\\ * *A$u+OPP<<& h - -CDNCSTUU ;;u  %%r7c`^\rSrSr%Sr/SQrS\S'/SQrS\S'S U4SjjrS Sjr S r U=r $) AdjustSaturationi%aAdjust color saturation of an image. The input image is expected to be an RGB image in the range of [0, 1]. Args: saturation_factor: How much to adjust the saturation. 0 will give a black and white image, 1 will give the original image while 2 will enhance the saturation by a factor of 2. saturation_mode: The mode to adjust saturation. Shape: - Input: Image/torch.Tensor to be adjusted in the shape of :math:`(*, 3, H, W)`. - Output: Adjusted image in the shape of :math:`(*, 3, H, W)`. Example: >>> x = torch.ones(1, 3, 3, 3) >>> AdjustSaturation(2.)(x) tensor([[[[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]], [[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]], [[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]]]]) >>> x = torch.ones(2, 3, 3, 3) >>> y = torch.ones(2) >>> out = AdjustSaturation(y)(x) >>> torch.nn.functional.mse_loss(x, out) tensor(0.) rMrrMrMClassVar[list[int]]ONNX_DEFAULT_INPUTSHAPEONNX_DEFAULT_OUTPUTSHAPEc.>[TU]5 XlgNsuper__init__saturation_factorselfr __class__s r5rAdjustSaturation.__init__M =Nr7c,[XR5$r)rArrrTs r5forwardAdjustSaturation.forwardQ (>(>??r7rrUnion[float, torch.Tensor]rNonerTrrr __name__ __module__ __qualname____firstlineno____doc__r__annotations__rrr__static_attributes__ __classcell__rs@r5rr%s6"H4C0B4C1CO@@r7rc`^\rSrSr%Sr/SQrS\S'/SQrS\S'S U4SjjrS Sjr S r U=r $) #AdjustSaturationWithGraySubtractioniUaAdjust color saturation of an image. This implementation aligns PIL. Hence, the output is close to TorchVision. The input image is expected to be in the range of [0, 1]. The input image is expected to be an RGB or gray image in the range of [0, 1]. Args: saturation_factor: How much to adjust the saturation. 0 will give a black and white image, 1 will give the original image while 2 will enhance the saturation by a factor of 2. saturation_mode: The mode to adjust saturation. Shape: - Input: Image/torch.Tensor to be adjusted in the shape of :math:`(*, 3, H, W)`. - Output: Adjusted image in the shape of :math:`(*, 3, H, W)`. Example: >>> x = torch.ones(1, 3, 3, 3) >>> AdjustSaturationWithGraySubtraction(2.)(x) tensor([[[[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]], [[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]], [[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]]]]) >>> x = torch.ones(2, 3, 3, 3) >>> y = torch.ones(2) >>> out = AdjustSaturationWithGraySubtraction(y)(x) >>> torch.nn.functional.mse_loss(x, out) tensor(0.) rrrrc.>[TU]5 Xlgrrrs r5r,AdjustSaturationWithGraySubtraction.__init__rr7c,[XR5$r)r=rrs r5r+AdjustSaturationWithGraySubtraction.forwards6u>T>TUUr7rrrrrs@r5r r Us6%N4C0B4C1COVVr7r c`^\rSrSr%Sr/SQrS\S'/SQrS\S'S U4SjjrS Sjr S r U=r $) AdjustHueiaAdjust hue of an image. This implementation aligns PIL. Hence, the output is close to TorchVision. The input image is expected to be in the range of [0, 1]. The input image is expected to be an RGB image in the range of [0, 1]. Args: hue_factor: How much to shift the hue channel. Should be in [-PI, PI]. PI and -PI give complete reversal of hue channel in HSV space in positive and negative direction respectively. 0 means no shift. Therefore, both -PI and PI will give an image with complementary colors while 0 gives the original image. Shape: - Input: Image/torch.Tensor to be adjusted in the shape of :math:`(*, 3, H, W)`. - Output: Adjusted image in the shape of :math:`(*, 3, H, W)`. Example: >>> x = torch.ones(1, 3, 3, 3) >>> AdjustHue(3.141516)(x) tensor([[[[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]], [[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]], [[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]]]]) >>> x = torch.ones(2, 3, 3, 3) >>> y = torch.ones(2) * 3.141516 >>> AdjustHue(y)(x).shape torch.Size([2, 3, 3, 3]) rrrrc.>[TU]5 Xlgr)rr hue_factor)rrrs r5rAdjustHue.__init__s 6@r7c,[XR5$r)rJrrs r5rAdjustHue.forwards%11r7)r)rrrrrrrs@r5rrs4%N4C0B4C1CA22r7rc>^\rSrSrSrSSU4SjjjrSSjrSrU=r$) AdjustGammaiaPerform gamma correction on an image. The input image is expected to be in the range of [0, 1]. Args: gamma: Non negative real number, same as y\gammay in the equation. gamma larger than 1 make the shadows darker, while gamma smaller than 1 make dark regions lighter. gain: The constant multiplier. Shape: - Input: Image to be adjusted in the shape of :math:`(*, N)`. - Output: Adjusted image in the shape of :math:`(*, N)`. Example: >>> x = torch.ones(1, 1, 3, 3) >>> AdjustGamma(1.0, 2.0)(x) tensor([[[[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]]]]) >>> x = torch.ones(2, 5, 3, 3) >>> y1 = torch.ones(2) * 1.0 >>> y2 = torch.ones(2) * 2.0 >>> AdjustGamma(y1, y2)(x).shape torch.Size([2, 5, 3, 3]) c:>[TU]5 XlX lgr)rrrUrV)rrUrVrs r5rAdjustGamma.__init__s 16 04 r7cB[XRUR5$r)rYrUrVrs r5rAdjustGamma.forwardsE::tyy99r7)rVrUr:)rUrrVrrrr rrrrrrrrrrs@r5rrs:55 ::r7rc:^\rSrSrSrSU4SjjrSSjrSrU=r$)AdjustContrastiatAdjust Contrast of an image. This implementation aligns OpenCV, not PIL. Hence, the output differs from TorchVision. The input image is expected to be in the range of [0, 1]. Args: contrast_factor: Contrast adjust factor per element in the batch. 0 generates a completely black image, 1 does not modify the input image while any other non-negative number modify the brightness by this factor. Shape: - Input: Image/Input to be adjusted in the shape of :math:`(*, N)`. - Output: Adjusted image in the shape of :math:`(*, N)`. Example: >>> x = torch.ones(1, 1, 3, 3) >>> AdjustContrast(0.5)(x) tensor([[[[0.5000, 0.5000, 0.5000], [0.5000, 0.5000, 0.5000], [0.5000, 0.5000, 0.5000]]]]) >>> x = torch.ones(2, 5, 3, 3) >>> y = torch.ones(2) >>> AdjustContrast(y)(x).shape torch.Size([2, 5, 3, 3]) c.>[TU]5 Xlgrrrcontrast_factorrr"rs r5rAdjustContrast.__init__ ;Jr7c,[XR5$r)r^r"rs r5rAdjustContrast.forwardsu&:&:;;r7r"r"rrrrrrs@r5rrs:K<>> x = torch.ones(1, 1, 3, 3) >>> AdjustContrastWithMeanSubtraction(0.5)(x) tensor([[[[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]]]]) >>> x = torch.ones(2, 5, 3, 3) >>> y = torch.ones(2) >>> AdjustContrastWithMeanSubtraction(y)(x).shape torch.Size([2, 5, 3, 3]) c.>[TU]5 Xlgrr!r#s r5r*AdjustContrastWithMeanSubtraction.__init__'r%r7c,[XR5$r)rcr"rs r5r)AdjustContrastWithMeanSubtraction.forward+s4U>> x = torch.ones(1, 1, 3, 3) >>> AdjustBrightness(1.)(x) tensor([[[[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]]]]) >>> x = torch.ones(2, 5, 3, 3) >>> y = torch.ones(2) >>> AdjustBrightness(y)(x).shape torch.Size([2, 5, 3, 3]) c.>[TU]5 Xlgrrrbrightness_factorrr4rs r5rAdjustBrightness.__init__Lrr7c,[XR5$r)rfr4rs r5rAdjustBrightness.forwardPrr7r4r4rrrrrrs@r5r1r1/s8O@@r7r1c>^\rSrSrSrSSU4SjjjrSSjrSrU=r$) AdjustSigmoidiTaAdjust the contrast of an image torch.Tensor or performs sigmoid correction on the input image torch.Tensor. The input image is expected to be in the range of [0, 1]. Reference: [1]: Gustav J. Braun, "Image Lightness Rescaling Using Sigmoidal Contrast Enhancement Functions", http://markfairchild.org/PDFs/PAP07.pdf Args: image: Image to be adjusted in the shape of :math:`(*, H, W)`. cutoff: The cutoff of sigmoid function. gain: The multiplier of sigmoid function. inv: If is set to True the function will return the negative sigmoid correction. Example: >>> x = torch.ones(1, 1, 2, 2) >>> AdjustSigmoid(gain=0)(x) tensor([[[[0.5000, 0.5000], [0.5000, 0.5000]]]]) cF>[TU]5 XlX lX0lgr)rrrkrVrl)rrkrVrlrs r5rAdjustSigmoid.__init__ks #  r7cT[XRURURS9$)NrkrVrl)rmrkrVrlrr.s r5rAdjustSigmoid.forwardqseKKdiiTXXVVr7r@r} F)rkr$rVr$rlboolrrr.rrrrrs@r5r<r<Ts, WWr7r<c>^\rSrSrSrSSU4SjjjrSSjrSrU=r$) AdjustLogiuanAdjust log correction on the input image torch.Tensor. The input image is expected to be in the range of [0, 1]. Reference: [1]: http://www.ece.ucsb.edu/Faculty/Manjunath/courses/ece178W03/EnhancePart1.pdf Args: image: Image to be adjusted in the shape of :math:`(*, H, W)`. gain: The multiplier of logarithmic function. inv: If is set to True the function will return the inverse logarithmic correction. clip_output: Whether to clip the output image with range of [0, 1]. Example: >>> x = torch.zeros(1, 1, 2, 2) >>> AdjustLog(inv=True)(x) tensor([[[[0., 0.], [0., 0.]]]]) cF>[TU]5 XlX lX0lgr)rrrVrlr\)rrVrlr\rs r5rAdjustLog.__init__s  !,r7cT[XRURURS9$)N)rVrlr\)rprVrlr\rAs r5rAdjustLog.forwards%iiTXX4K[K[\\r7)r\rVrlrFT)rVr$rlrEr\rErrrFrrs@r5rHrHus*-- ]]r7rHc:^\rSrSrSrSU4SjjrSSjrSrU=r$)AdjustBrightnessAccumulativeia+Adjust Brightness of an image accumulatively. This implementation aligns PIL. Hence, the output is close to TorchVision. The input image is expected to be in the range of [0, 1]. Args: brightness_factor: Brightness adjust factor per element in the batch. 0 does not modify the input image while any other number modify the brightness. Shape: - Input: Image/Input to be adjusted in the shape of :math:`(*, N)`. - Output: Adjusted image in the shape of :math:`(*, N)`. Example: >>> x = torch.ones(1, 1, 3, 3) >>> AdjustBrightnessAccumulative(1.)(x) tensor([[[[1., 1., 1.], [1., 1., 1.], [1., 1., 1.]]]]) >>> x = torch.ones(2, 5, 3, 3) >>> y = torch.ones(2) >>> AdjustBrightnessAccumulative(y)(x).shape torch.Size([2, 5, 3, 3]) c.>[TU]5 Xlgrr3r5s r5r%AdjustBrightnessAccumulative.__init__rr7c,[XR5$r)rhr4rs r5r$AdjustBrightnessAccumulative.forwards-e5K5KLLr7r9r:rrrs@r5rOrOs8OMMr7rOc>^\rSrSrSrSSU4SjjjrSSjrSrU=r$) InvertiaInvert the values of an input torch.Tensor by its maximum value. Args: input: The input torch.Tensor to invert with an arbitatry shape. max_val: The expected maximum value in the input torch.Tensor. The shape has to according to the input torch.Tensor shape, or at least has to work with broadcasting. Default: 1.0. Example: >>> img = torch.rand(1, 2, 4, 4) >>> Invert()(img).shape torch.Size([1, 2, 4, 4]) >>> img = 255. * torch.rand(1, 2, 3, 4, 4) >>> Invert(torch.as_tensor(255.))(img).shape torch.Size([1, 2, 3, 4, 4]) >>> img = torch.rand(1, 3, 4, 4) >>> Invert(torch.as_tensor([[[[1.]]]]))(img).shape torch.Size([1, 3, 4, 4]) c>[TU]5 Uc[R"S5n[ U[ R 5(dURSU5 gXlg)Nr:r) rrr%rr#r Parameterregister_bufferr)rrrs r5rInvert.__init__sG  ?ll3'G'2<<00  G 4"Lr7c,[XR5$r)rrrs r5rInvert.forwardse\\**r7)rr)rOptional[torch.Tensor]rrrrrs@r5rUrUs,##++r7rU)r.rr/rrrr)rTrrUrrVrrr)T)r.rr/rr\rErrrC) r.rrkr$rVr$rlrErrrM) r.rrVr$rlrEr\rErr)r})rTrryrrr)r}N)rTrryrrz$Optional[Union[float, torch.Tensor]]rr)rTrrzUnion[int, torch.Tensor]rr)rTrr/rrr)rrrrr/rrr)rrrrrr)rrrrrr)r.rrr\rr)< __future__rmathrtypingrrrr%torch.nn.functionalrrr kornia.colorr r r kornia.core.checkr r rkornia.core.utilsrkornia.image.utilsrrr6r=rArHrJrYr^rcrfrhrmrpr{rrrrrrrrrModulerr rrrr+r1r<rHrOrUr7r5rgso$#,, AA 0Q<6r'TB(X`cG G :GB\GGTAH7vRVC C!;CJNCCNRV4 4!;4JN44nD#L?B.16:B( B(*B(4B( B(JV9V9r4a4an"< $'T:."&J-@ryy-@`0V"))0Vf02 02f$:"))$:N#