def roi_feature_transform(self, blobs_in, rpn_ret, blob_rois='rois', method='RoIPoolF',
resolution=7, spatial_scale=1. / 16., sampling_ratio=0, panet=False):
"""Add the specified RoI pooling method. The sampling_ratio argument
is supported for some, but not all, RoI transform methods.
RoIFeatureTransform abstracts away:
- Use of FPN or not
- Specifics of the transform method
"""
assert method in {'RoIPoolF', 'RoICrop', 'RoIAlign'}, \
'Unknown pooling method: {}'.format(method)
if isinstance(blobs_in, list):
# FPN case: add RoIFeatureTransform to each FPN level
device_id = blobs_in[0].get_device()
k_max = cfg.FPN.ROI_MAX_LEVEL # coarsest level of pyramid
k_min = cfg.FPN.ROI_MIN_LEVEL # finest level of pyramid
assert len(blobs_in) == k_max - k_min + 1
bl_out_list = []
for lvl in range(k_min, k_max + 1):
bl_in = blobs_in[k_max - lvl] # blobs_in is in reversed order
sc = spatial_scale[k_max - lvl] # in reversed order
if not panet:
bl_rois = blob_rois + '_fpn' + str(lvl)
else:
bl_rois = blob_rois
if len(rpn_ret[bl_rois]):
rois = Variable(torch.from_numpy(rpn_ret[bl_rois])).cuda(device_id)
if method == 'RoIPoolF':
# Warning!: Not check if implementation matches Detectron
xform_out = RoIPoolFunction(resolution, resolution, sc)(bl_in, rois)
elif method == 'RoICrop':
# Warning!: Not check if implementation matches Detectron
grid_xy = net_utils.affine_grid_gen(
rois, bl_in.size()[2:], self.grid_size)
grid_yx = torch.stack(
[grid_xy.data[:, :, :, 1], grid_xy.data[:, :, :, 0]], 3).contiguous()
xform_out = RoICropFunction()(bl_in, Variable(grid_yx).detach())
if cfg.CROP_RESIZE_WITH_MAX_POOL:
xform_out = F.max_pool2d(xform_out, 2, 2)
elif method == 'RoIAlign':
xform_out = RoIAlignFunction(
resolution, resolution, sc, sampling_ratio)(bl_in, rois)
bl_out_list.append(xform_out)
if not panet:
# The pooled features from all levels are concatenated along the
# batch dimension into a single 4D tensor.
xform_shuffled = torch.cat(bl_out_list, dim=0)
# Unshuffle to match rois from dataloader
device_id = xform_shuffled.get_device()
restore_bl = rpn_ret[blob_rois + '_idx_restore_int32']
restore_bl = Variable(
torch.from_numpy(restore_bl.astype('int64', copy=False))).cuda(device_id)
xform_out = xform_shuffled[restore_bl]
else:
return bl_out_list
else:
# Single feature level
# rois: holds R regions of interest, each is a 5-tuple
# (batch_idx, x1, y1, x2, y2) specifying an image batch index and a
# rectangle (x1, y1, x2, y2)
device_id = blobs_in.get_device()
rois = Variable(torch.from_numpy(rpn_ret[blob_rois])).cuda(device_id)
if method == 'RoIPoolF':
xform_out = RoIPoolFunction(resolution, resolution, spatial_scale)(blobs_in, rois)
elif method == 'RoICrop':
grid_xy = net_utils.affine_grid_gen(rois, blobs_in.size()[2:], self.grid_size)
grid_yx = torch.stack(
[grid_xy.data[:, :, :, 1], grid_xy.data[:, :, :, 0]], 3).contiguous()
xform_out = RoICropFunction()(blobs_in, Variable(grid_yx).detach())
if cfg.CROP_RESIZE_WITH_MAX_POOL:
xform_out = F.max_pool2d(xform_out, 2, 2)
elif method == 'RoIAlign':
xform_out = RoIAlignFunction(
resolution, resolution, spatial_scale, sampling_ratio)(blobs_in, rois)
return xform_out
def roi_feature_transform(self, blobs_in, rpn_ret, blob_rois='rois', method='RoIPoolF',
resolution=7, spatial_scale=1. / 16., sampling_ratio=0):
"""Add the specified RoI pooling method. The sampling_ratio argument
is supported for some, but not all, RoI transform methods.
RoIFeatureTransform abstracts away:
- Use of FPN or not
- Specifics of the transform method
"""
assert method in {'RoIPoolF', 'RoICrop', 'RoIAlign'}, \
'Unknown pooling method: {}'.format(method)
if isinstance(blobs_in, list):
# FPN case: add RoIFeatureTransform to each FPN level
device_id = blobs_in[0].get_device()
k_max = cfg.FPN.ROI_MAX_LEVEL # coarsest level of pyramid
k_min = cfg.FPN.ROI_MIN_LEVEL # finest level of pyramid
assert len(blobs_in) == k_max - k_min + 1
bl_out_list = []
for lvl in range(k_min, k_max + 1):
bl_in = blobs_in[k_max - lvl] # blobs_in is in reversed order
sc = spatial_scale[k_max - lvl] # in reversed order
bl_rois = blob_rois + '_fpn' + str(lvl)
if len(rpn_ret[bl_rois]):
rois = Variable(torch.from_numpy(rpn_ret[bl_rois])).cuda(device_id)
if method == 'RoIPoolF':
# Warning!: Not check if implementation matches Detectron
xform_out = RoIPoolFunction(resolution, resolution, sc)(bl_in, rois)
elif method == 'RoICrop':
# Warning!: Not check if implementation matches Detectron
grid_xy = net_utils.affine_grid_gen(
rois, bl_in.size()[2:], self.grid_size)
grid_yx = torch.stack(
[grid_xy.data[:, :, :, 1], grid_xy.data[:, :, :, 0]], 3).contiguous()
xform_out = RoICropFunction()(bl_in, Variable(grid_yx).detach())
if cfg.CROP_RESIZE_WITH_MAX_POOL:
xform_out = F.max_pool2d(xform_out, 2, 2)
elif method == 'RoIAlign':
xform_out = RoIAlignFunction(
resolution, resolution, sc, sampling_ratio)(bl_in, rois)
bl_out_list.append(xform_out)
# The pooled features from all levels are concatenated along the
# batch dimension into a single 4D tensor.
xform_shuffled = torch.cat(bl_out_list, dim=0)
# Unshuffle to match rois from dataloader
device_id = xform_shuffled.get_device()
restore_bl = rpn_ret[blob_rois + '_idx_restore_int32']
restore_bl = Variable(
torch.from_numpy(restore_bl.astype('int64', copy=False))).cuda(device_id)
xform_out = xform_shuffled[restore_bl]
else:
# Single feature level
# rois: holds R regions of interest, each is a 5-tuple
# (batch_idx, x1, y1, x2, y2) specifying an image batch index and a
# rectangle (x1, y1, x2, y2)
device_id = blobs_in.get_device()
rois = Variable(torch.from_numpy(rpn_ret[blob_rois])).cuda(device_id)
if method == 'RoIPoolF':
xform_out = RoIPoolFunction(resolution, resolution, spatial_scale)(blobs_in, rois)
elif method == 'RoICrop':
grid_xy = net_utils.affine_grid_gen(rois, blobs_in.size()[2:], self.grid_size)
grid_yx = torch.stack(
[grid_xy.data[:, :, :, 1], grid_xy.data[:, :, :, 0]], 3).contiguous()
xform_out = RoICropFunction()(blobs_in, Variable(grid_yx).detach())
if cfg.CROP_RESIZE_WITH_MAX_POOL:
xform_out = F.max_pool2d(xform_out, 2, 2)
elif method == 'RoIAlign':
xform_out = RoIAlignFunction(
resolution, resolution, spatial_scale, sampling_ratio)(blobs_in, rois)
return xform_out
def roi_feature_transform(self,
blobs_in,
rpn_ret,
blob_rois='rois',
method='RoIPoolF',
resolution=7,
spatial_scale=1. / 16.,
sampling_ratio=0):
"""Add the specified RoI pooling method. The sampling_ratio argument
is supported for some, but not all, RoI transform methods.
RoIFeatureTransform abstracts away:
- Use of FPN or not
- Specifics of the transform method
"""
assert method in {'RoIPoolF', 'RoICrop', 'RoIAlign'}, \
'Unknown pooling method: {}'.format(method)
if isinstance(blobs_in, list):
# FPN case: add RoIFeatureTransform to each FPN level
device_id = blobs_in[0].get_device()
k_max = cfg.FPN.ROI_MAX_LEVEL # coarsest level of pyramid
k_min = cfg.FPN.ROI_MIN_LEVEL # finest level of pyramid
assert len(blobs_in) == k_max - k_min + 1
bl_out_list = []
for lvl in range(k_min, k_max + 1):
bl_in = blobs_in[k_max - lvl] # blobs_in is in reversed order
sc = spatial_scale[k_max - lvl] # in reversed order
bl_rois = blob_rois + '_fpn' + str(lvl)
if len(rpn_ret[bl_rois]):
rois = Variable(torch.from_numpy(
rpn_ret[bl_rois])).cuda(device_id)
if method == 'RoIPoolF':
# Warning!: Not check if implementation matches Detectron
xform_out = RoIPoolFunction(resolution, resolution,
sc)(bl_in, rois)
elif method == 'RoICrop':
# Warning!: Not check if implementation matches Detectron
grid_xy = net_utils.affine_grid_gen(
rois,
bl_in.size()[2:], self.grid_size)
grid_yx = torch.stack([
grid_xy.data[:, :, :, 1], grid_xy.data[:, :, :, 0]
], 3).contiguous()
xform_out = RoICropFunction()(
bl_in, Variable(grid_yx).detach())
if cfg.CROP_RESIZE_WITH_MAX_POOL:
xform_out = F.max_pool2d(xform_out, 2, 2)
elif method == 'RoIAlign':
xform_out = RoIAlignFunction(resolution, resolution,
sc, sampling_ratio)(bl_in,
rois)
# print(xform_out.size())
# input()
bl_out_list.append(xform_out)
#print(xform_out.size())
#input()
#if bl_out_list[0].size(2) == 7:
# FLAG_MASK = False
#else:
# FLAG_MASK = True
#feats = torch.FloatTensor(bl_out_list[0].size()).fill_(0).cuda(device_id)
#feats = torch.autograd.Variable(feats)
#if FLAG_MASK:
# for i in range(len(bl_out_list)):
# feats_t = bl_out_list[i]
# feats_t = F.relu(self.mask_fcn_1[i](feats_t))
# feats = torch.max(feats, feats_t)
#else:
# for i in range(len(bl_out_list)):
# feats_t = bl_out_list[i]
# feats_t = F.relu(self.box_fcn_1[i](feats_t))
# feats = torch.max(feats, feats_t)
#
## The pooled features from all levels are concatenated along the
## batch dimension into a single 4D tensor.
## xform_shuffled = torch.cat(bl_out_list, dim=0)
#xform_shuffled = feats
#
## Unshuffle to match rois from dataloader
#device_id = xform_shuffled.get_device()
#xform_out = xform_shuffled
#print(xform_out.size())
#else:
# # Single feature level
# # rois: holds R regions of interest, each is a 5-tuple
# # (batch_idx, x1, y1, x2, y2) specifying an image batch index and a
# # rectangle (x1, y1, x2, y2)
# device_id = blobs_in.get_device()
# rois = Variable(torch.from_numpy(rpn_ret[blob_rois])).cuda(device_id)
# if method == 'RoIPoolF':
# xform_out = RoIPoolFunction(resolution, resolution, spatial_scale)(blobs_in, rois)
# elif method == 'RoICrop':
# grid_xy = net_utils.affine_grid_gen(rois, blobs_in.size()[2:], self.grid_size)
# grid_yx = torch.stack(
# [grid_xy.data[:, :, :, 1], grid_xy.data[:, :, :, 0]], 3).contiguous()
# xform_out = RoICropFunction()(blobs_in, Variable(grid_yx).detach())
# if cfg.CROP_RESIZE_WITH_MAX_POOL:
# xform_out = F.max_pool2d(xform_out, 2, 2)
# elif method == 'RoIAlign':
# xform_out = RoIAlignFunction(
# resolution, resolution, spatial_scale, sampling_ratio)(blobs_in, rois)
return bl_out_list