def AffNetHardNet_describeFromKeys(img_np, KPlist):
img = torch.autograd.Variable(torch.from_numpy(img_np.astype(np.float32)),
volatile=True)
img = img.view(1, 1, img.size(0), img.size(1))
HessianAffine = ScaleSpaceAffinePatchExtractor(mrSize=5.192,
num_features=0,
border=0,
num_Baum_iters=0)
if USE_CUDA:
HessianAffine = HessianAffine.cuda()
img = img.cuda()
with torch.no_grad():
HessianAffine.createScaleSpace(
img) # to generate scale pyramids and stuff
descriptors = []
Alist = []
n = 0
# for patch_np in patches:
for kp in KPlist:
x, y = np.float32(kp.pt)
LAFs = normalizeLAFs(
torch.tensor([[AffNetPix.PS / 2, 0, x], [0, AffNetPix.PS / 2,
y]]).reshape(1, 2, 3),
img.size(3), img.size(2))
with torch.no_grad():
patch = HessianAffine.extract_patches_from_pyr(denormalizeLAFs(
LAFs, img.size(3), img.size(2)),
PS=AffNetPix.PS)
if WRITE_IMGS_DEBUG:
SaveImageWithKeys(patch.detach().cpu().numpy().reshape([32, 32]),
[], 'p2/' + str(n) + '.png')
if USE_CUDA:
# or ---> A = AffNetPix(subpatches.cuda()).cpu()
with torch.no_grad():
A = batched_forward(AffNetPix, patch.cuda(), 256).cpu()
else:
with torch.no_grad():
A = AffNetPix(patch)
new_LAFs = torch.cat([torch.bmm(A, LAFs[:, :, 0:2]), LAFs[:, :, 2:]],
dim=2)
dLAFs = denormalizeLAFs(new_LAFs, img.size(3), img.size(2))
with torch.no_grad():
patchaff = HessianAffine.extract_patches_from_pyr(dLAFs, PS=32)
if WRITE_IMGS_DEBUG:
SaveImageWithKeys(
patchaff.detach().cpu().numpy().reshape([32, 32]), [],
'p1/' + str(n) + '.png')
SaveImageWithKeys(img_np, [kp], 'im1/' + str(n) + '.png')
descriptors.append(
HardNetDescriptor(patchaff).cpu().numpy().astype(np.float32))
Alist.append(
convertLAFs_to_A23format(LAFs.detach().cpu().numpy().astype(
np.float32)))
n = n + 1
return descriptors, Alist
def forward(self,x, random_Baum = False, random_resp = False, return_patches = False):
### Detection
num_features_prefilter = self.num
#if self.num_Baum_iters > 0:
# num_features_prefilter = 2 * self.num;
if random_resp:
num_features_prefilter *= 4
responses, LAFs, final_pyr_idxs, final_level_idxs, scale_pyr = self.multiScaleDetector(x,num_features_prefilter)
if random_resp:
if self.num < responses.size(0):
ridxs = torch.randperm(responses.size(0))[:self.num]
if x.is_cuda:
ridxs = ridxs.cuda()
responses = responses[ridxs]
LAFs = LAFs[ridxs ,:,:]
final_pyr_idxs = final_pyr_idxs[ridxs]
final_level_idxs = final_level_idxs[ridxs]
LAFs[:,0:2,0:2] = self.mrSize * LAFs[:,:,0:2]
n_iters = self.num_Baum_iters;
if random_Baum and (n_iters > 1):
n_iters = int(np.random.randint(1,n_iters + 1))
if n_iters > 0:
responses, LAFs, final_pyr_idxs, final_level_idxs = self.getAffineShape(scale_pyr, responses, LAFs,
final_pyr_idxs, final_level_idxs, self.num, n_iters = n_iters)
#LAFs = self.getOrientation(scale_pyr, LAFs, final_pyr_idxs, final_level_idxs)
#if return_patches:
# pyr_inv_idxs = get_inverted_pyr_index(scale_pyr, final_pyr_idxs, final_level_idxs)
# patches = extract_patches_from_pyramid_with_inv_index(scale_pyr, pyr_inv_idxs, LAFs, PS = self.PS)
if return_patches:
patches = extract_patches(x, LAFs, PS = self.PS)
else:
patches = None
return denormalizeLAFs(LAFs, x.size(3), x.size(2)), patches, responses#, scale_pyr
def getAffmaps_from_Affnet(patches_np):
sp1, sp2 = np.shape(patches_np[0])
subpatches = torch.autograd.Variable(
torch.zeros([len(patches_np), 1, 32, 32], dtype=torch.float32),
volatile=True).view(len(patches_np), 1, 32, 32)
for k in range(0, len(patches_np)):
subpatch = patches_np[k][int(sp1 / 2) - 16:int(sp2 / 2) + 16,
int(sp1 / 2) - 16:int(sp2 / 2) + 16].reshape(
1, 1, 32, 32)
subpatches[k, :, :, :] = torch.from_numpy(subpatch.astype(
np.float32)) #=subpatch
x, y = subpatches.shape[3] / 2.0 + 2, subpatches.shape[2] / 2.0 + 2
LAFs = normalizeLAFs(
torch.tensor([[AffNetPix.PS / 2, 0, x], [0, AffNetPix.PS / 2,
y]]).reshape(1, 2, 3),
subpatches.shape[3], subpatches.shape[2])
baseLAFs = torch.zeros([subpatches.shape[0], 2, 3], dtype=torch.float32)
for m in range(subpatches.shape[0]):
baseLAFs[m, :, :] = LAFs
if USE_CUDA:
# or ---> A = AffNetPix(subpatches.cuda()).cpu()
with torch.no_grad():
A = batched_forward(AffNetPix, subpatches.cuda(), 256).cpu()
else:
with torch.no_grad():
A = AffNetPix(subpatches)
LAFs = torch.cat([torch.bmm(A, baseLAFs[:, :, 0:2]), baseLAFs[:, :, 2:]],
dim=2)
dLAFs = denormalizeLAFs(LAFs, subpatches.shape[3], subpatches.shape[2])
Alist = convertLAFs_to_A23format(dLAFs.detach().cpu().numpy().astype(
np.float32))
return Alist
def forward(self, x, do_ori=False, verb=False):
### Detection
t = time.time()
num_features_prefilter = self.num
if self.num_Baum_iters > 0:
num_features_prefilter = int(1.5 * self.num)
responses, LAFs, final_pyr_idxs, final_level_idxs = self.multiScaleDetector(
x, num_features_prefilter)
if verb:
print((time.time() - t, 'detection multiscale'))
t = time.time()
LAFs[:, 0:2, 0:2] = self.mrSize * LAFs[:, :, 0:2]
if self.num_Baum_iters > 0:
responses, LAFs, final_pyr_idxs, final_level_idxs = self.getAffineShape(
responses, LAFs, final_pyr_idxs, final_level_idxs, self.num)
if verb:
print((time.time() - t, 'affine shape iters'))
t = time.time()
if do_ori:
LAFs = self.getOrientation(LAFs, final_pyr_idxs, final_level_idxs)
#pyr_inv_idxs = get_inverted_pyr_index(self.scale_pyr, final_pyr_idxs, final_level_idxs)
#patches = extract_patches_from_pyramid_with_inv_index(scale_pyr, pyr_inv_idxs, LAFs, PS = self.PS)
#patches = extract_patches(x, LAFs, PS = self.PS)
#print time.time() - t, len(LAFs), ' patches extraction'
return denormalizeLAFs(LAFs, x.size(3), x.size(2)), responses
def forward(self,x, do_ori = True):
### Detection
t = time.time()
num_features_prefilter = self.num
responses, LAFs, final_pyr_idxs, final_level_idxs = self.multiScaleDetectorAff(x,num_features_prefilter)
print time.time() - t, 'detection multiscale'
t = time.time()
LAFs[:,0:2,0:2] = self.mrSize * LAFs[:,:,0:2]
if do_ori:
LAFs = self.getOrientation(LAFs, final_pyr_idxs, final_level_idxs)
#pyr_inv_idxs = get_inverted_pyr_index(scale_pyr, final_pyr_idxs, final_level_idxs)
#patches = extract_patches_from_pyramid_with_inv_index(scale_pyr, pyr_inv_idxs, LAFs, PS = self.PS)
#patches = extract_patches(x, LAFs, PS = self.PS)
#print time.time() - t, len(LAFs), ' patches extraction'
return denormalizeLAFs(LAFs, x.size(3), x.size(2)), responses
def forward(self,
x,
random_Baum=False,
random_resp=False,
return_patches=False):
### Detection
num_features_prefilter = self.num
#if self.num_Baum_iters > 0:
# num_features_prefilter = 2 * self.num;
if random_resp:
num_features_prefilter *= 4
responses, LAFs, final_pyr_idxs, final_level_idxs, scale_pyr = self.multiScaleDetector(
x, num_features_prefilter)
if random_resp:
if self.num < responses.size(0):
ridxs = torch.randperm(responses.size(0))[:self.num]
if x.is_cuda:
ridxs = ridxs.cuda()
responses = responses[ridxs]
LAFs = LAFs[ridxs, :, :]
final_pyr_idxs = final_pyr_idxs[ridxs]
final_level_idxs = final_level_idxs[ridxs]
LAFs[:, 0:2, 0:2] = self.mrSize * LAFs[:, :, 0:2]
n_iters = self.num_Baum_iters
if random_Baum and (n_iters > 1):
n_iters = int(np.random.randint(1, n_iters + 1))
if n_iters > 0:
responses, LAFs, final_pyr_idxs, final_level_idxs, dets, A = self.ImproveLAFsEstimation(
scale_pyr,
responses,
LAFs,
final_pyr_idxs,
final_level_idxs,
self.num,
n_iters=n_iters)
if return_patches:
patches = extract_patches(x, LAFs, PS=self.PS)
else:
patches = None
return denormalizeLAFs(
LAFs, x.size(3),
x.size(2)), patches, responses, dets, A #, scale_pyr
def AffNetHardNet_describe(patches):
descriptors = np.zeros(shape=[patches.shape[0], 128], dtype=np.float32)
HessianAffine = []
subpatches = torch.autograd.Variable(torch.zeros([len(patches), 1, 32, 32],
dtype=torch.float32),
volatile=True).view(
len(patches), 1, 32, 32)
baseLAFs = torch.zeros([len(patches), 2, 3], dtype=torch.float32)
for m in range(patches.shape[0]):
patch_np = patches[m, :, :, 0].reshape(np.shape(patches)[1:3])
HessianAffine.append(
ScaleSpaceAffinePatchExtractor(mrSize=5.192,
num_features=0,
border=0,
num_Baum_iters=0))
with torch.no_grad():
var_image = torch.autograd.Variable(torch.from_numpy(
patch_np.astype(np.float32)),
volatile=True)
patch = var_image.view(1, 1, var_image.size(0), var_image.size(1))
with torch.no_grad():
HessianAffine[m].createScaleSpace(
patch) # to generate scale pyramids and stuff
x, y = patch.size(3) / 2.0 + 2, patch.size(2) / 2.0 + 2
LAFs = normalizeLAFs(
torch.tensor([[AffNetPix.PS / 2, 0, x], [0, AffNetPix.PS / 2,
y]]).reshape(1, 2, 3),
patch.size(3), patch.size(2))
baseLAFs[m, :, :] = LAFs
with torch.no_grad():
subpatch = HessianAffine[m].extract_patches_from_pyr(
denormalizeLAFs(LAFs, patch.size(3), patch.size(2)),
PS=AffNetPix.PS)
if WRITE_IMGS_DEBUG:
SaveImageWithKeys(
subpatch.detach().cpu().numpy().reshape([32, 32]), [],
'p1/' + str(n) + '.png')
# This subpatch has been blured by extract_patches _from_pyr...
# let't us crop it manually to obtain fair results agains other methods
subpatch = patch_np[16:48, 16:48].reshape(1, 1, 32, 32)
#var_image = torch.autograd.Variable(torch.from_numpy(subpatch.astype(np.float32)), volatile = True)
#subpatch = var_image.view(1, 1, 32,32)
subpatches[m, :, :, :] = torch.from_numpy(
subpatch.astype(np.float32)) #=subpatch
if WRITE_IMGS_DEBUG:
SaveImageWithKeys(
subpatch.detach().cpu().numpy().reshape([32, 32]), [],
'p2/' + str(n) + '.png')
if USE_CUDA:
# or ---> A = AffNetPix(subpatches.cuda()).cpu()
with torch.no_grad():
A = batched_forward(AffNetPix, subpatches.cuda(), 256).cpu()
else:
with torch.no_grad():
A = AffNetPix(subpatches)
LAFs = torch.cat([torch.bmm(A, baseLAFs[:, :, 0:2]), baseLAFs[:, :, 2:]],
dim=2)
dLAFs = denormalizeLAFs(LAFs, patch.size(3), patch.size(2))
Alist = convertLAFs_to_A23format(dLAFs.detach().cpu().numpy().astype(
np.float32))
for m in range(patches.shape[0]):
with torch.no_grad():
patchaff = HessianAffine[m].extract_patches_from_pyr(
dLAFs[m, :, :].reshape(1, 2, 3), PS=32)
if WRITE_IMGS_DEBUG:
SaveImageWithKeys(
patchaff.detach().cpu().numpy().reshape([32, 32]), [],
'im1/' + str(n) + '.png')
SaveImageWithKeys(patch_np, [], 'im2/' + str(n) + '.png')
subpatches[m, :, :, :] = patchaff
if USE_CUDA:
with torch.no_grad():
# descriptors = HardNetDescriptor(subpatches.cuda()).detach().cpu().numpy().astype(np.float32)
descriptors = batched_forward(HardNetDescriptor, subpatches.cuda(),
256).cpu().numpy().astype(np.float32)
else:
with torch.no_grad():
descriptors = HardNetDescriptor(
subpatches).detach().cpu().numpy().astype(np.float32)
return descriptors, Alist
abc = self.features(self.input_norm(input))
return abc2A(abc[:, 0, :, :].contiguous() + 1.,
abc[:, 1, :, :].contiguous(),
abc[:, 2, :, :].contiguous() + 1.)
def weights_init(m):
if isinstance(m, nn.Conv2d):
nn.init.orthogonal(m.weight.data, gain=1.0)
try:
nn.init.constant(m.bias.data, 0.01)
except:
pass
return
HA = ScaleSpaceAffinePatchExtractor(mrSize=5.0,
num_features=3000,
border=1,
num_Baum_iters=5,
AffNet=BaumNet())
LAFs, patches, resp, pyr = HA(var_image_reshape / 255.)
LAFs = denormalizeLAFs(LAFs, img.shape[1], img.shape[0],
use_cuda=USE_CUDA).data.cpu().numpy()
ells = LAFs2ell(LAFs)
np.savetxt(output_fname, ells, delimiter=' ', fmt='%10.10f')
#line_prepender(output_fname, str(len(ells)))
#line_prepender(output_fname, '1.0')