a d/@s ddlmZmZddlZddlmZddlmZmZddl m Z dej eej e fe eej ej ddd Zdej e eej ej d d d Zdej e eej ej d d dZej ej dddZGdddejZGdddejZGdddejZGdddejZdS))OptionalUnionN)gaussian_blur2dspatial_gradientKORNIA_CHECK_SHAPE{Gz?sobel)inputk grads_modesigmasreturnc Cst|gd|durft|tjs4tdt|t|jdkrV|d|dkrft d|jt ||}|dddddf}|dddddf}t |ddd }t |ddd }t ||dd } ||| | } ||} | || d} |dur| | d  d ddd} | S) u(Compute the Harris cornerness function. 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) 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 r9Invalid sigmas shape, we expect B == input.size(0). Got: r?r)r isinstancetorchTensor TypeErrortypelenshapesize ValueErrorrrpowview) r r r r gradientsdxdydx2dy2dxydet_mtrace_mscoresr2a/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/kornia/feature/responses.pyharris_response s$9 "  r4)r r r rcCst|gdt||}|dddddf}|dddddf}t|ddd}t|ddd}t||dd}||||} ||} d| t| dd | } d| t| dd | } t| | } |dur| |d d ddd} | S) aCompute the Shi-Tomasi cornerness function. Function does not do any normalization or nms. The response map is computed according the following formulation: .. math:: R = min(eig(M)) 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]]]]) rNrrrrrg?rr) rrrrsqrtabsminr'r()r r r r)r*r+r,r-r.r/r0e1e2r1r2r2r3 gftt_responseas2 "" r:cCst|gd|durft|tjs4tdt|t|jdkrV|d|dkrft d|jt ||d}|dddddf}|dddddf}|dddddf}|||d}|dur|| d d ddd}|S) aCompute the absolute of determinant of the Hessian matrix. Function does not do any normalization or nms. The response map is computed according the following formulation: .. math:: R = det(H) 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]]]]) rNrrrrrrr) rrrr r!r"r#r$r%r&rr'r()r r r r)Zdxxr.Zdyyr1r2r2r3hessian_responses6 " r;)r rcCsBt|gd|ddddddf|ddddddfS)zCompute the Difference-of-Gaussian response. Args: input: a given the gaussian 5d tensor :math:`(B, C, D, H, W)`. Return: the response map per channel with shape :math:`(B, C, D-1, H, W)`. )rrLrrNrrr)r r2r2r3 dog_responses r=csReZdZdZddfdd ZedddZd eje ejejdd d Z Z S) BlobDoGzuModule that calculates Difference-of-Gaussians blobs. See :func:`~kornia.feature.dog_response` for details. NrcstdSN)super__init__self __class__r2r3rB s zBlobDoG.__init__cCs|jjSr@)rF__name__rCr2r2r3__repr__szBlobDoG.__repr__r r rcCst|Sr@)r=rDr r r2r2r3forwardszBlobDoG.forward)N rG __module__ __qualname____doc__rBstrrHrr rrK __classcell__r2r2rEr3r>sr>cs`eZdZdZd eeejfddfdd Ze ddd Z deje ejejd d d Z Z S) CornerHarrisziModule that calculates Harris corners. See :func:`~kornia.feature.harris_response` for details. r N)r rcs@tt|tur*|dt|n |d|||_dS)Nr )rArBr"floatZregister_bufferrZtensorr )rDr r rEr2r3rBs    zCornerHarris.__init__r?cCs(|jjdt|jdd|jdS)Nz(k=z, grads_mode=))rFrGrPr r rCr2r2r3rH%szCornerHarris.__repr__rIcCst||j|j|Sr@)r4r r rJr2r2r3rK(szCornerHarris.forward)r )N)rGrMrNrOrrSrr rBrPrHrrKrQr2r2rEr3rRs rRcsTeZdZdZd ddfdd ZedddZd eje ejejd d d Z Z S) CornerGFTTzkModule that calculates Shi-Tomasi corners. See :func:`~kornia.feature.gfft_response` for details. r Nr?cst||_dSr@rArBr rDr rEr2r3rB2s zCornerGFTT.__init__cCs|jjd|jdSNrTrUrFrGr rCr2r2r3rH7szCornerGFTT.__repr__rIcCst||j|Sr@)r:r rJr2r2r3rK:szCornerGFTT.forward)r )NrLr2r2rEr3rV,srVcsTeZdZdZd ddfdd ZedddZd eje ejejd d d Z Z S) BlobHessianziModule that calculates Hessian blobs. See :func:`~kornia.feature.hessian_response` for details. r Nr?cst||_dSr@rWrXrEr2r3rBDs zBlobHessian.__init__cCs|jjd|jdSrYrZrCr2r2r3rHIszBlobHessian.__repr__rIcCst||j|Sr@)r;r rJr2r2r3rKLszBlobHessian.forward)r )NrLr2r2rEr3r[>sr[)rr N)r N)r N)typingrrrZtorch.nnnnZkornia.filtersrrZkornia.testingrr rSrPr4r:r;r=Moduler>rRrVr[r2r2r2r3s:    X K L