Python mmdet.ops.MaskedConv2d() Examples
The following are 30
code examples of mmdet.ops.MaskedConv2d().
You can vote up the ones you like or vote down the ones you don't like,
and go to the original project or source file by following the links above each example.
You may also want to check out all available functions/classes of the module
mmdet.ops
, or try the search function
.
.
Example #1
| Source File: guided_anchor_head.py From RDSNet with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #2
| Source File: guided_anchor_head.py From ttfnet with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.in_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.in_channels, self.num_anchors * 2, 1)
self.feature_adaption = FeatureAdaption(
self.in_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #3
| Source File: guided_anchor_head.py From CenterNet with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #4
| Source File: guided_anchor_head.py From Feature-Selective-Anchor-Free-Module-for-Single-Shot-Object-Detection with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #5
| Source File: guided_anchor_head.py From Cascade-RPN with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #6
| Source File: guided_anchor_head.py From FoveaBox with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #7
| Source File: guided_anchor_head.py From Libra_R-CNN with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #8
| Source File: guided_anchor_head.py From IoU-Uniform-R-CNN with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.in_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.in_channels, self.num_anchors * 2, 1)
self.feature_adaption = FeatureAdaption(
self.in_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #9
| Source File: guided_anchor_head.py From kaggle-kuzushiji-recognition with MIT License | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #10
| Source File: guided_anchor_head.py From PolarMask with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #11
| Source File: guided_anchor_head.py From mmdetection_with_SENet154 with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #12
| Source File: guided_anchor_head.py From mmdetection with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.in_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.in_channels, self.num_anchors * 2, 1)
self.feature_adaption = FeatureAdaption(
self.in_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #13
| Source File: guided_anchor_head.py From AerialDetection with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #14
| Source File: guided_anchor_head.py From GCNet with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #15
| Source File: guided_anchor_head.py From mmdetection-annotated with Apache License 2.0 | 5 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.conv_cls = MaskedConv2d(self.feat_channels,
self.num_anchors * self.cls_out_channels,
1)
self.conv_reg = MaskedConv2d(self.feat_channels, self.num_anchors * 4,
1)
Example #16
| Source File: ga_retina_head.py From mmdetection with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
"""Initialize layers of the head."""
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(
self.feat_channels,
self.num_anchors * self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(
self.feat_channels, self.num_anchors * 4, 3, padding=1)
Example #17
| Source File: ga_retina_head.py From IoU-Uniform-R-CNN with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(
self.feat_channels,
self.num_anchors * self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(
self.feat_channels, self.num_anchors * 4, 3, padding=1)
Example #18
| Source File: ga_retina_head.py From RDSNet with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(
self.feat_channels,
self.num_anchors * self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(
self.feat_channels, self.num_anchors * 4, 3, padding=1)
Example #19
| Source File: ga_retina_head.py From Libra_R-CNN with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(self.feat_channels,
self.num_anchors *
self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(self.feat_channels,
self.num_anchors * 4,
3,
padding=1)
Example #20
| Source File: ga_retina_head.py From mmdetection-annotated with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(self.feat_channels,
self.num_anchors *
self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(self.feat_channels,
self.num_anchors * 4,
3,
padding=1)
Example #21
| Source File: ga_retina_head.py From FoveaBox with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(
self.feat_channels,
self.num_anchors * self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(
self.feat_channels, self.num_anchors * 4, 3, padding=1)
Example #22
| Source File: ga_retina_head.py From mmdetection_with_SENet154 with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(self.feat_channels,
self.num_anchors *
self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(self.feat_channels,
self.num_anchors * 4,
3,
padding=1)
Example #23
| Source File: ga_retina_head.py From Cascade-RPN with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(
self.feat_channels,
self.num_anchors * self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(
self.feat_channels, self.num_anchors * 4, 3, padding=1)
Example #24
| Source File: ga_retina_head.py From GCNet with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(self.feat_channels,
self.num_anchors *
self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(self.feat_channels,
self.num_anchors * 4,
3,
padding=1)
Example #25
| Source File: ga_retina_head.py From Feature-Selective-Anchor-Free-Module-for-Single-Shot-Object-Detection with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(
self.feat_channels,
self.num_anchors * self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(
self.feat_channels, self.num_anchors * 4, 3, padding=1)
Example #26
| Source File: ga_retina_head.py From kaggle-kuzushiji-recognition with MIT License | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(
self.feat_channels,
self.num_anchors * self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(
self.feat_channels, self.num_anchors * 4, 3, padding=1)
Example #27
| Source File: ga_retina_head.py From CenterNet with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(self.feat_channels,
self.num_anchors *
self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(self.feat_channels,
self.num_anchors * 4,
3,
padding=1)
Example #28
| Source File: ga_retina_head.py From AerialDetection with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(self.feat_channels,
self.num_anchors *
self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(self.feat_channels,
self.num_anchors * 4,
3,
padding=1)
Example #29
| Source File: ga_retina_head.py From ttfnet with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(
self.feat_channels,
self.num_anchors * self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(
self.feat_channels, self.num_anchors * 4, 3, padding=1)
Example #30
| Source File: ga_retina_head.py From PolarMask with Apache License 2.0 | 4 votes |
|
def _init_layers(self):
self.relu = nn.ReLU(inplace=True)
self.cls_convs = nn.ModuleList()
self.reg_convs = nn.ModuleList()
for i in range(self.stacked_convs):
chn = self.in_channels if i == 0 else self.feat_channels
self.cls_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.reg_convs.append(
ConvModule(
chn,
self.feat_channels,
3,
stride=1,
padding=1,
conv_cfg=self.conv_cfg,
norm_cfg=self.norm_cfg))
self.conv_loc = nn.Conv2d(self.feat_channels, 1, 1)
self.conv_shape = nn.Conv2d(self.feat_channels, self.num_anchors * 2,
1)
self.feature_adaption_cls = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.feature_adaption_reg = FeatureAdaption(
self.feat_channels,
self.feat_channels,
kernel_size=3,
deformable_groups=self.deformable_groups)
self.retina_cls = MaskedConv2d(
self.feat_channels,
self.num_anchors * self.cls_out_channels,
3,
padding=1)
self.retina_reg = MaskedConv2d(
self.feat_channels, self.num_anchors * 4, 3, padding=1)
