def forward(self, x):
x_shape = x.size() # (b, c, h, w)
offset = self.offset_filter(x) # (b, 2*c, h, w)
offset_w, offset_h = torch.split(offset, self.regular_filter.in_channels, 1) # (b, c, h, w)
offset_w = offset_w.contiguous().view(-1, int(x_shape[2]), int(x_shape[3])) # (b*c, h, w)
offset_h = offset_h.contiguous().view(-1, int(x_shape[2]), int(x_shape[3])) # (b*c, h, w)
if not self.input_shape or self.input_shape != x_shape:
self.input_shape = x_shape
grid_w, grid_h = np.meshgrid(np.linspace(-1, 1, x_shape[3]), np.linspace(-1, 1, x_shape[2])) # (h, w)
grid_w = torch.Tensor(grid_w)
grid_h = torch.Tensor(grid_h)
if self.cuda:
grid_w = grid_w.cuda()
grid_h = grid_h.cuda()
self.grid_w = nn.Parameter(grid_w)
self.grid_h = nn.Parameter(grid_h)
offset_w = offset_w + self.grid_w # (b*c, h, w)
offset_h = offset_h + self.grid_h # (b*c, h, w)
x = x.contiguous().view(-1, int(x_shape[2]), int(x_shape[3])).unsqueeze(1) # (b*c, 1, h, w)
x = F.grid_sample(x, torch.stack((offset_h, offset_w), 3)) # (b*c, h, w)
x = x.contiguous().view(-1, int(x_shape[1]), int(x_shape[2]), int(x_shape[3])) # (b, c, h, w)
x = self.regular_filter(x)
return x
python类grid_sample()的实例源码
def _crop_pool_layer(self, bottom, rois, max_pool=True):
# implement it using stn
# box to affine
# input (x1,y1,x2,y2)
"""
[ x2-x1 x1 + x2 - W + 1 ]
[ ----- 0 --------------- ]
[ W - 1 W - 1 ]
[ ]
[ y2-y1 y1 + y2 - H + 1 ]
[ 0 ----- --------------- ]
[ H - 1 H - 1 ]
"""
rois = rois.detach()
x1 = rois[:, 1::4] / 16.0
y1 = rois[:, 2::4] / 16.0
x2 = rois[:, 3::4] / 16.0
y2 = rois[:, 4::4] / 16.0
height = bottom.size(2)
width = bottom.size(3)
# affine theta
theta = Variable(rois.data.new(rois.size(0), 2, 3).zero_())
theta[:, 0, 0] = (x2 - x1) / (width - 1)
theta[:, 0 ,2] = (x1 + x2 - width + 1) / (width - 1)
theta[:, 1, 1] = (y2 - y1) / (height - 1)
theta[:, 1, 2] = (y1 + y2 - height + 1) / (height - 1)
if max_pool:
pre_pool_size = cfg.POOLING_SIZE * 2
grid = F.affine_grid(theta, torch.Size((rois.size(0), 1, pre_pool_size, pre_pool_size)))
crops = F.grid_sample(bottom.expand(rois.size(0), bottom.size(1), bottom.size(2), bottom.size(3)), grid)
crops = F.max_pool2d(crops, 2, 2)
else:
grid = F.affine_grid(theta, torch.Size((rois.size(0), 1, cfg.POOLING_SIZE, cfg.POOLING_SIZE)))
crops = F.grid_sample(bottom.expand(rois.size(0), bottom.size(1), bottom.size(2), bottom.size(3)), grid)
return crops
def window_to_image(z_where, window_size, image_size, windows):
n = windows.size(0)
assert windows.size(1) == window_size ** 2, 'Size mismatch.'
theta = expand_z_where(z_where)
grid = F.affine_grid(theta, torch.Size((n, 1, image_size, image_size)))
out = F.grid_sample(windows.view(n, 1, window_size, window_size), grid)
return out.view(n, image_size, image_size)
def image_to_window(z_where, window_size, image_size, images):
n = images.size(0)
assert images.size(1) == images.size(2) == image_size, 'Size mismatch.'
theta_inv = expand_z_where(z_where_inv(z_where))
grid = F.affine_grid(theta_inv, torch.Size((n, 1, window_size, window_size)))
out = F.grid_sample(images.view(n, 1, image_size, image_size), grid)
return out.view(n, -1)
# Helper to expand parameters to the size of the mini-batch. I would
# like to remove this and just write `t.expand(n, -1)` inline, but the
# `-1` argument of `expand` doesn't seem to work with PyTorch 0.2.0.
def compare_grid_sample():
# do gradcheck
N = random.randint(1, 8)
C = 2 # random.randint(1, 8)
H = 5 # random.randint(1, 8)
W = 4 # random.randint(1, 8)
input = Variable(torch.randn(N, C, H, W).cuda(), requires_grad=True)
input_p = input.clone().data.contiguous()
grid = Variable(torch.randn(N, H, W, 2).cuda(), requires_grad=True)
grid_clone = grid.clone().contiguous()
out_offcial = F.grid_sample(input, grid)
grad_outputs = Variable(torch.rand(out_offcial.size()).cuda())
grad_outputs_clone = grad_outputs.clone().contiguous()
grad_inputs = torch.autograd.grad(out_offcial, (input, grid), grad_outputs.contiguous())
grad_input_off = grad_inputs[0]
crf = RoICropFunction()
grid_yx = torch.stack([grid_clone.data[:,:,:,1], grid_clone.data[:,:,:,0]], 3).contiguous().cuda()
out_stn = crf.forward(input_p, grid_yx)
grad_inputs = crf.backward(grad_outputs_clone.data)
grad_input_stn = grad_inputs[0]
pdb.set_trace()
delta = (grad_input_off.data - grad_input_stn).sum()
def stn(self, x):
xs = self.localization(x)
xs = xs.view(-1, 10 * 3 * 3)
theta = self.fc_loc(xs)
theta = theta.view(-1, 2, 3)
grid = F.affine_grid(theta, x.size())
x = F.grid_sample(x, grid)
return x
def _crop_pool_layer(bottom, rois, max_pool=True):
# code modified from
# https://github.com/ruotianluo/pytorch-faster-rcnn
# implement it using stn
# box to affine
# input (x1,y1,x2,y2)
"""
[ x2-x1 x1 + x2 - W + 1 ]
[ ----- 0 --------------- ]
[ W - 1 W - 1 ]
[ ]
[ y2-y1 y1 + y2 - H + 1 ]
[ 0 ----- --------------- ]
[ H - 1 H - 1 ]
"""
rois = rois.detach()
batch_size = bottom.size(0)
D = bottom.size(1)
H = bottom.size(2)
W = bottom.size(3)
roi_per_batch = rois.size(0) / batch_size
x1 = rois[:, 1::4] / 16.0
y1 = rois[:, 2::4] / 16.0
x2 = rois[:, 3::4] / 16.0
y2 = rois[:, 4::4] / 16.0
height = bottom.size(2)
width = bottom.size(3)
# affine theta
zero = Variable(rois.data.new(rois.size(0), 1).zero_())
theta = torch.cat([\
(x2 - x1) / (width - 1),
zero,
(x1 + x2 - width + 1) / (width - 1),
zero,
(y2 - y1) / (height - 1),
(y1 + y2 - height + 1) / (height - 1)], 1).view(-1, 2, 3)
if max_pool:
pre_pool_size = cfg.POOLING_SIZE * 2
grid = F.affine_grid(theta, torch.Size((rois.size(0), 1, pre_pool_size, pre_pool_size)))
bottom = bottom.view(1, batch_size, D, H, W).contiguous().expand(roi_per_batch, batch_size, D, H, W)\
.contiguous().view(-1, D, H, W)
crops = F.grid_sample(bottom, grid)
crops = F.max_pool2d(crops, 2, 2)
else:
grid = F.affine_grid(theta, torch.Size((rois.size(0), 1, cfg.POOLING_SIZE, cfg.POOLING_SIZE)))
bottom = bottom.view(1, batch_size, D, H, W).contiguous().expand(roi_per_batch, batch_size, D, H, W)\
.contiguous().view(-1, D, H, W)
crops = F.grid_sample(bottom, grid)
return crops, grid
