3j9 SSKJrJr SSKrSSKJr SSKJr SSKJrJ r S\ S\ 4Sjr SS \RS \\R\ 4S \S \\RS\R4 S jjrSS \RS \S \\RS\R4SjjrSS \RS \S \\RS\R4SjjrS \RS\R4SjrS S \RS\ S\ S\R4Sjjr"SS\R(5r"SS\R(5r"SS\R(5r"SS\R(5r"SS\R(5rg)!)OptionalUnionN)nnKORNIA_CHECK_SHAPE)gaussian_blur2dspatial_gradientsigmareturncD[SU-S-5nUS-S:XaUS- nU$)Ng @?r)int)r ksizes R/home/wildlama/miniconda3/lib/python3.13/site-packages/kornia/feature/responses.py_get_kernel_sizers1  E!C' (E  qyA~   Linputk grads_modesigmasc`[U/SQ5 Ub[U[R5(d[ S[ U535e[ UR5S:Xa$URS5URS5:wa[SUR35e[X5nUSS2SS2S4nUSS2SS2S4n[US-SS 5n[US-SS 5n[XV-SS 5n Xx-X-- n Xx-n XU S--- n Ub%XRS 5RS SSS5-n U $) uCompute the Harris cornerness function. .. image:: _static/img/harris_response.png Function does not do any normalization or nms. The response map is computed according the following formulation: .. math:: R = max(0, det(M) - k \cdot trace(M)^2) torch.where: .. math:: M = \sum_{(x,y) \in W} \begin{bmatrix} I^{2}_x & I_x I_y \\ I_x I_y & I^{2}_y \\ \end{bmatrix} and :math:`k` is an empirically determined constant :math:`k ∈ [ 0.04 , 0.06 ]` Args: input: input image with shape :math:`(B, C, H, W)`. k: the Harris detector free parameter. grads_mode: can be ``'sobel'`` for standalone use or ``'diff'`` for use on Gaussian pyramid. sigmas: coefficients to be multiplied by multichannel response. Should be shape of :math:`(B)` It is necessary for performing non-maxima-suppression across different scale pyramid levels. See `vlfeat `_. Return: the response map per channel with shape :math:`(B, C, H, W)`. Example: >>> input = torch.tensor([[[ ... [0., 0., 0., 0., 0., 0., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 0., 0., 0., 0., 0., 0.], ... ]]]) # 1x1x7x7 >>> # compute the response map harris_response(input, 0.04) tensor([[[[0.0012, 0.0039, 0.0020, 0.0000, 0.0020, 0.0039, 0.0012], [0.0039, 0.0065, 0.0040, 0.0000, 0.0040, 0.0065, 0.0039], [0.0020, 0.0040, 0.0029, 0.0000, 0.0029, 0.0040, 0.0020], [0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000], [0.0020, 0.0040, 0.0029, 0.0000, 0.0029, 0.0040, 0.0020], [0.0039, 0.0065, 0.0040, 0.0000, 0.0040, 0.0065, 0.0039], [0.0012, 0.0039, 0.0020, 0.0000, 0.0020, 0.0039, 0.0012]]]]) BCHWN'sigmas type is not a torch.Tensor. Got rr9Invalid sigmas shape, we expect B == input.size(0). Got: rr"r r )r isinstancetorchTensor TypeErrortypelenshapesize ValueErrorr rpowview) rrrr gradientsdxdydx2dy2dxydet_mtrace_mscoress rharris_responser:'s5xu23 &%,,//Ed6l^TU UFLL!Q&FKKNejjm,KXY_YeYeXfgh h.uAI Aq)B Aq)B(AvzBC'AvzBC'DC)ci/EIG! #33F **Q-,,RAq99 Mrc([U/SQ5 [X5nUSS2SS2S4nUSS2SS2S4n[US-SS5n[US-SS5n[XE-SS5nXg-X-- n Xg-n SU [R"U S-S U -- R 55--n SU [R"U S-S U -- R 55- -n [R "X5n Ub%XRS 5RS SSS5-n U $) aCompute the Shi-Tomasi cornerness function. .. image:: _static/img/gftt_response.png Function does not do any normalization or nms. The response map is computed according the following formulation: .. math:: R = min(eig(M)) torch.where: .. math:: M = \sum_{(x,y) \in W} \begin{bmatrix} I^{2}_x & I_x I_y \\ I_x I_y & I^{2}_y \\ \end{bmatrix} Args: input: input image with shape :math:`(B, C, H, W)`. grads_mode: can be ``'sobel'`` for standalone use or ``'diff'`` for use on Gaussian pyramid. sigmas: coefficients to be multiplied by multichannel response. Should be shape of :math:`(B)` It is necessary for performing non-maxima-suppression across different scale pyramid levels. See `vlfeat `_. Return: the response map per channel with shape :math:`(B, C, H, W)`. Example: >>> input = torch.tensor([[[ ... [0., 0., 0., 0., 0., 0., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 0., 0., 0., 0., 0., 0.], ... ]]]) # 1x1x7x7 >>> # compute the response map gftt_response(input) tensor([[[[0.0155, 0.0334, 0.0194, 0.0000, 0.0194, 0.0334, 0.0155], [0.0334, 0.0575, 0.0339, 0.0000, 0.0339, 0.0575, 0.0334], [0.0194, 0.0339, 0.0497, 0.0000, 0.0497, 0.0339, 0.0194], [0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000], [0.0194, 0.0339, 0.0497, 0.0000, 0.0497, 0.0339, 0.0194], [0.0334, 0.0575, 0.0339, 0.0000, 0.0339, 0.0575, 0.0334], [0.0155, 0.0334, 0.0194, 0.0000, 0.0194, 0.0334, 0.0155]]]]) rNrrrr!r#g?r$r%) rr rr'sqrtabsminr/r0)rrrr1r2r3r4r5r6r7r8e1e2r9s r gftt_responserAsju23.uAI Aq)B Aq)B'AvzBC'AvzBC'DC)ci/EIGg GQJU4J3O3O3Q(RRSBg GQJU4J3O3O3Q(RRSB 99R,F **Q-,,RAq99 Mrc[U/SQ5 Ub[U[R5(d[ S[ U535e[ UR5S:Xa$URS5URS5:wa[SUR35e[XS5nUSS2SS2S4nUSS2SS2S4nUSS2SS2S4nXF-US-- nUb%XrRS5RS SSS5-nU$) asCompute the absolute of determinant of the Hessian matrix. .. image:: _static/img/hessian_response.png Function does not do any normalization or nms. The response map is computed according the following formulation: .. math:: R = det(H) torch.where: .. math:: M = \sum_{(x,y) \in W} \begin{bmatrix} I_{xx} & I_{xy} \\ I_{xy} & I_{yy} \\ \end{bmatrix} Args: input: input image with shape :math:`(B, C, H, W)`. grads_mode: can be ``'sobel'`` for standalone use or ``'diff'`` for use on Gaussian pyramid. sigmas: coefficients to be multiplied by multichannel response. Should be shape of :math:`(B)` It is necessary for performing non-maxima-suppression across different scale pyramid levels. See `vlfeat `_. Return: the response map per channel with shape :math:`(B, C, H, W)`. Shape: - Input: :math:`(B, C, H, W)` - Output: :math:`(B, C, H, W)` Examples: >>> input = torch.tensor([[[ ... [0., 0., 0., 0., 0., 0., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 1., 1., 1., 1., 1., 0.], ... [0., 0., 0., 0., 0., 0., 0.], ... ]]]) # 1x1x7x7 >>> # compute the response map hessian_response(input) tensor([[[[0.0155, 0.0334, 0.0194, 0.0000, 0.0194, 0.0334, 0.0155], [0.0334, 0.0575, 0.0339, 0.0000, 0.0339, 0.0575, 0.0334], [0.0194, 0.0339, 0.0497, 0.0000, 0.0497, 0.0339, 0.0194], [0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000], [0.0194, 0.0339, 0.0497, 0.0000, 0.0497, 0.0339, 0.0194], [0.0334, 0.0575, 0.0339, 0.0000, 0.0339, 0.0575, 0.0334], [0.0155, 0.0334, 0.0194, 0.0000, 0.0194, 0.0334, 0.0155]]]]) rNrrrr rr$r%) rr&r'r(r)r*r+r,r-r.r r/r0)rrrr1dxxr6dyyr9s rhessian_responserEsru23 &%,,//Ed6l^TU UFLL!Q&FKKNejjm,KXY_YeYeXfgh h.u!DI!!Q'*C!!Q'*C!!Q'*C9sAv-F **Q-,,RAq99 MrcX[U/SQ5 USS2SS2SS24USS2SS2SS24- $)zCompute the Difference-of-Gaussian response. Args: input: a given the gaussian 5d torch.Tensor :math:`(B, C, D, H, W)`. Return: the response map per channel with shape :math:`(B, C, D-1, H, W)`. )rrLrrNrr%r)rs r dog_responserHs5u78 Aqr?U1a"9- --rsigma1sigma2c[U/SQ5 [U5n[U5n[XU4X45n[XU4X"45nXe- $)a9Compute the Difference-of-Gaussian response. .. image:: _static/img/dog_response_single.png Args: input: a given the gaussian 4d torch.Tensor :math:`(B, C, H, W)`. sigma1: lower gaussian sigma sigma2: bigger gaussian sigma Return: the response map per channel with shape :math:`(B, C, H, W)`. r)rrr)rrIrJks1ks2g1g2s rdog_response_singlerP,sPu23 6 "C 6 "C c V,< =B c V,< =B 7Nrc^\rSrSrSrS U4SjjrS\4SjrS S\RS\ \RS\R4Sjjr S r U=r $) BlobDoGiBzqnn.Module that calculates Difference-of-Gaussians blobs. See :func: `~kornia.feature.dog_response` for details. r c">[TU]5 gN)super__init__)self __class__s rrVBlobDoG.__init__Is rc.URR$rT)rX__name__rWs r__repr__BlobDoG.__repr__Ls~~&&&rrrc[U5$rT)rHrWrrs rforwardBlobDoG.forwardOs E""r)r NrT)r[ __module__ __qualname____firstlineno____doc__rVstrr]r'r(rra__static_attributes__ __classcell__rXs@rrRrRBsL '#'#U\\#8ELL3I#UZUaUa##rrRc^\rSrSrSrS S\S\SS4U4SjjjrS\4SjrSS \ RS \ \ RS\ R4S jjr S r U=r$) BlobDoGSingleiSznn.Module that calculates Difference-of-Gaussians blobs. .. image:: _static/img/dog_response_single.png See :func:`~kornia.feature.dog_response_single` for details. rIrJr Nc:>[TU]5 XlX lgrT)rUrVrIrJ)rWrIrJrXs rrVBlobDoGSingle.__init__[s   rchURRSURSURS3$)Nz , sigma1=z , sigma2=))rXr[rIrJr\s rr]BlobDoGSingle.__repr__`s/..))*)DKK= $++VWXXrrrcB[XRUR5$rT)rPrIrJr`s rraBlobDoGSingle.forwardcs"5++t{{CCr)rIrJr g?rT)r[rdrerfrgfloatrVrhr]r'r(rrarirjrks@rrmrmSsmuED Y#YDU\\D8ELL3IDUZUaUaDDrrmc^\rSrSr%Sr\R \S'S S\\ \R 4S\ SS4U4Sjjjr S\ 4Sjr SS \R S \ \R S\R 4S jjrS rU=r$) CornerHarrisigznn.Module that calculates Harris corners. .. image:: _static/img/harris_response.png See :func:`~kornia.feature.harris_response` for details. rrr Nc>[TU]5 [U[5(a'UR S[ R "U55 OUR SU5 X lg)Nr)rUrVr&rvregister_bufferr'tensorr)rWrrrXs rrVCornerHarris.__init__qsH  a    ell1o 6  a ()rchURRSURSURS3$)Nz(k=z , grads_mode=rq)rXr[rrr\s rr]CornerHarris.__repr__ys0..))*#dffX]4??BSSTUUrrrcD[XRURU5$rT)r:rrr`s rraCornerHarris.forward|suffdoovFFrrsobelrT)r[rdrerfrgr'r(__annotations__rrvrhrVr]rrarirjrks@rrxrxgs ||O*%u|| 34*#*TX**V#VGU\\G8ELL3IGUZUaUaGGrrxc^\rSrSrSrS S\SS4U4SjjjrS\4SjrS S\RS \ \RS\R4S jjr S r U=r $) CornerGFTTiznn.Module that calculates Shi-Tomasi corners. .. image:: _static/img/gftt_response.png See :func:`~kornia.feature.gftt_response` for details. rr Nc.>[TU]5 XlgrTrUrVrrWrrXs rrVCornerGFTT.__init__ )rcNURRSURS3$Nz (grads_mode=rqrXr[rr\s rr]CornerGFTT.__repr__%..))*,t6GqIIrrrc.[XRU5$rT)rArr`s rraCornerGFTT.forwardsUOOV<[TU]5 XlgrTrrs rrVBlobHessian.__init__rrcNURRSURS3$rrr\s rr]BlobHessian.__repr__rrrrc.[XRU5$rT)rErr`s rraBlobHessian.forwards??rrrrTrrks@rrrsf*3*T**J#J@U\\@8ELL3I@UZUaUa@@rr)g{Gz?rN)rNru)typingrrr'rkornia.core.checkrkornia.filtersrr rvrrr(rhr:rArErHrPModulerRrmrxrrrcrrrs$# 0< E c %)%) W <<W U\\5 !WW U\\ " W  \\ WvVZJ <<J%(J