a d@sdZddlmZddlZddlmmZddlm Z ddl m Z ddZ e dfejejejd d d Zdejeejd ddZdejejeejdddZdejejeeefedddZdS)zImplementation of "differentiable spatial to numerical" (soft-argmax) operations, as described in the paper "Numerical Coordinate Regression with Convolutional Neural Networks" by Nibali et al.)TupleN)check_is_tensor)create_meshgridcCs*t|t|jdks&td|jdS)Nz-Invalid input shape, we expect BxCxHxW. Got: )rlenshape ValueError)tensorr d/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/kornia/geometry/subpix/dsnt.py$_validate_batched_image_tensor_input sr g?)input temperaturereturncCsXt||j\}}}}|j|j|jd}|||d}tj||dd}|||||S)a1Apply the Softmax function over features in each image channel. Note that this function behaves differently to :py:class:`torch.nn.Softmax2d`, which instead applies Softmax over features at each spatial location. Args: input: the input tensor with shape :math:`(B, N, H, W)`. temperature: factor to apply to input, adjusting the "smoothness" of the output distribution. Returns: a 2D probability distribution per image channel with shape :math:`(B, N, H, W)`. Examples: >>> heatmaps = torch.tensor([[[ ... [0., 0., 0.], ... [0., 0., 0.], ... [0., 1., 2.]]]]) >>> spatial_softmax2d(heatmaps) tensor([[[[0.0585, 0.0585, 0.0585], [0.0585, 0.0585, 0.0585], [0.0585, 0.1589, 0.4319]]]]) )devicedtype)Zdim)r rtorrviewFZsoftmax)r r batch_sizechannelsheightwidthxZx_softr r r spatial_softmax2ds rT)r normalized_coordinatesrc Cst||j\}}}}t||||j}||j}|dd}|dd}|||d} tj || ddd} tj || ddd} t | | gd} | ||dS)aCompute the expectation of coordinate values using spatial probabilities. The input heatmap is assumed to represent a valid spatial probability distribution, which can be achieved using :func:`~kornia.geometry.subpixel.spatial_softmax2d`. Args: input: the input tensor representing dense spatial probabilities with shape :math:`(B, N, H, W)`. normalized_coordinates: whether to return the coordinates normalized in the range of :math:`[-1, 1]`. Otherwise, it will return the coordinates in the range of the input shape. Returns: expected value of the 2D coordinates with shape :math:`(B, N, 2)`. Output order of the coordinates is (x, y). Examples: >>> heatmaps = torch.tensor([[[ ... [0., 0., 0.], ... [0., 0., 0.], ... [0., 1., 0.]]]]) >>> spatial_expectation2d(heatmaps, False) tensor([[[1., 2.]]]) .rr.TZkeepdim) r rrrrrZreshapertorchsumcat) r rrrrrgridpos_xpos_yZ input_flatZ expected_yZ expected_xoutputr r r spatial_expectation2d5s r)3#I9) numerator denominatorepsrcCs|tj||dS)N)min)r"clamp)r+r,r-r r r _safe_zero_divisionasr0)meanstdsizercCs|j|jkr|j|jks td|\}}t||||j}||j}|d||}|d||}||dd} ||dd} dt|d} dt|d} t| | } t| | }| |}|j dd d j d d d }t ||}|S) aRender the PDF of a 2D Gaussian distribution. Args: mean: the mean location of the Gaussian to render, :math:`(\mu_x, \mu_y)`. Shape: :math:`(*, 2)`. std: the standard deviation of the Gaussian to render, :math:`(\sigma_x, \sigma_y)`. Shape :math:`(*, 2)`. Should be able to be broadcast with `mean`. size: the (height, width) of the output image. normalized_coordinates: whether ``mean`` and ``std`` are assumed to use coordinates normalized in the range of :math:`[-1, 1]`. Otherwise, coordinates are assumed to be in the range of the output shape. Returns: tensor including rendered points with shape :math:`(*, H, W)`. z1Expected inputs to have the same dtype and devicerr).rNNr!).rNNgTr r) rr TypeErrorrrrr"Z reciprocalexpr#r0)r1r2r3rrrr%r&r'Zdist_xZdist_yZk_xZk_yZexps_xZexps_ygaussZval_sumr r r render_gaussian2des"  r8)T)r*)T)__doc__typingrr"Ztorch.nn.functionalnnZ functionalrZkornia.testingrZkornia.utils.gridrr r ZTensorrboolr)floatr0intr8r r r r s   "",