Skip to content

GitLab

Commit 89022a26 authored by Jiaqi Wang's avatar Jiaqi Wang Committed by Kai Chen
Browse files

Code of CVPR 2019 Paper: Region Proposal by Guided Anchoring (#594) 

* add two stage w/o neck and w/ upperneck

* add rpn r50 c4

* update c4 configs

* fix

* config update

* update config

* minor update

* mask rcnn support c4 train and test

* lr fix

* cascade support upper_neck

* add cascade c4 config

* update config

* update

* update res_layer to new interface

* refactoring

* c4 configs update

* refactoring

* update rpn_c4 config

* rename upper_neck as shared_head

* update

* update configs

* update

* update c4 configs

* update according to commits

* update

* add ga rpn

* test bug fix

* test bug fix with loc_filter_thr is large

* update configs

* update configs

* add ga_retinanet

* ga test bug fix

* update configs

* update

* init masked conv

* update

* update masked conv

* update

* support no ga_sampler

* update

* update

* test with masked_conv

* update comment

* fix flake errors
...
parent 3cb84acc
Hide whitespace changes
Inline Side-by-side
Showing with 2075 additions and 19 deletions
MODEL_ZOO.md
View file @ 89022a26
...@@ -214,6 +214,10 @@ Please refer to [Weight Standardization](configs/gn+ws/README.md) for details. ...@@ -214,6 +214,10 @@ Please refer to [Weight Standardization](configs/gn+ws/README.md) for details.
Please refer to [Deformable Convolutional Networks](configs/dcn/README.md) for details. Please refer to [Deformable Convolutional Networks](configs/dcn/README.md) for details.
### Guided Anchoring
Please refer to [Guided Anchoring](configs/guided_anchoring/README.md) for details.
## Comparison with Detectron and maskrcnn-benchmark ## Comparison with Detectron and maskrcnn-benchmark
......
compile.sh
View file @ 89022a26
...@@ -36,3 +36,10 @@ if [ -d "build" ]; then ...@@ -36,3 +36,10 @@ if [ -d "build" ]; then
rm -r build rm -r build
fi fi
$PYTHON setup.py build_ext --inplace $PYTHON setup.py build_ext --inplace
echo "Building masked conv op..."
cd ../masked_conv
if [ -d "build" ]; then
rm -r build
fi
$PYTHON setup.py build_ext --inplace
configs/guided_anchoring/README.md 0 → 100644
View file @ 89022a26
# Region Proposal by Guided Anchoring
## Introduction
We provide config files to reproduce the results in the CVPR 2019 paper for [Region Proposal by Guided Anchoring](https://arxiv.org/abs/1901.03278).
```
@inproceedings{wang2019region,
title={Region Proposal by Guided Anchoring},
author={Jiaqi Wang and Kai Chen and Shuo Yang and Chen Change Loy and Dahua Lin},
booktitle={IEEE Conference on Computer Vision and Pattern Recognition},
year={2019}
}
```
## Results and Models
The results on COCO 2017 val is shown in the below table. (results on test-dev are usually slightly higher than val).
| Method | Backbone | Style | Lr schd | Mem (GB) | Train time (s/iter) | Inf time (fps) | AR 1000 | Download |
| :----: | :-------------: | :-----: | :-----: | :------: | :-----------------: | :------------: | :-----: | :-------------------------------------------------------------------------------------------------------------------------------------------: |
| GA-RPN | R-50-FPN | caffe | 1x | 5.0 | 0.55 | 13.3 | 68.5 | [model](https://s3.ap-northeast-2.amazonaws.com/open-mmlab/mmdetection/models/guided_anchoring/ga_rpn_r50_caffe_fpn_1x_20190513-95e91886.pth) |
| GA-RPN | R-101-FPN | caffe | 1x | - | - | - | 69.6 | - |
| GA-RPN | X-101-32x4d-FPN | pytorch | 1x | - | - | - | 70.0 | - |
| GA-RPN | X-101-64x4d-FPN | pytorch | 1x | - | - | - | 70.5 | - |
| Method | Backbone | Style | Lr schd | Mem (GB) | Train time (s/iter) | Inf time (fps) | box AP | Download |
| :------------: | :-------------: | :-----: | :-----: | :------: | :-----------------: | :------------: | :----: | :-------------------------------------------------------------------------------------------------------------------------------------------------: |
| GA-Fast RCNN | R-50-FPN | caffe | 1x | 3.3 | 0.23 | 14.9 | 39.5 | [model](https://s3.ap-northeast-2.amazonaws.com/open-mmlab/mmdetection/models/guided_anchoring/ga_fast_r50_caffe_fpn_1x_20190513-c5af9f8b.pth) |
| GA-Faster RCNN | R-50-FPN | caffe | 1x | 5.1 | 0.64 | 9.6 | 39.9 | [model](https://s3.ap-northeast-2.amazonaws.com/open-mmlab/mmdetection/models/guided_anchoring/ga_faster_r50_caffe_fpn_1x_20190513-a52b31fa.pth) |
| GA-Faster RCNN | R-101-FPN | caffe | 1x | - | - | - | 41.5 | - |
| GA-Faster RCNN | X-101-32x4d-FPN | pytorch | 1x | - | - | - | 42.9 | - |
| GA-Faster RCNN | X-101-64x4d-FPN | pytorch | 1x | - | - | - | 43.9 | - |
| GA-RetinaNet | R-50-FPN | caffe | 1x | 3.2 | 0.50 | 10.7 | 37.0 | [model](https://s3.ap-northeast-2.amazonaws.com/open-mmlab/mmdetection/models/guided_anchoring/ga_retinanet_r50_caffe_fpn_1x_20190513-29905101.pth) |
| GA-RetinaNet | R-101-FPN | caffe | 1x | - | - | - | 38.9 | - |
| GA-RetinaNet | X-101-32x4d-FPN | pytorch | 1x | - | - | - | 40.3 | - |
| GA-RetinaNet | X-101-64x4d-FPN | pytorch | 1x | - | - | - | 40.8 | - |
- In the Guided Anchoring paper, `score_thr` is set to 0.001 in Fast/Faster RCNN and 0.05 in RetinaNet for both baselines and Guided Anchoring.
\ No newline at end of file
configs/guided_anchoring/ga_fast_r50_caffe_fpn_1x.py 0 → 100644
View file @ 89022a26
# model settings
model = dict(
type='FastRCNN',
pretrained='open-mmlab://resnet50_caffe',
backbone=dict(
type='ResNet',
depth=50,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
norm_cfg=dict(type='BN', requires_grad=False),
norm_eval=True,
style='caffe'),
neck=dict(
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=256,
num_outs=5),
bbox_roi_extractor=dict(
type='SingleRoIExtractor',
roi_layer=dict(type='RoIAlign', out_size=7, sample_num=2),
out_channels=256,
featmap_strides=[4, 8, 16, 32]),
bbox_head=dict(
type='SharedFCBBoxHead',
num_fcs=2,
in_channels=256,
fc_out_channels=1024,
roi_feat_size=7,
num_classes=81,
target_means=[0., 0., 0., 0.],
target_stds=[0.05, 0.05, 0.1, 0.1],
reg_class_agnostic=False,
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))
# model training and testing settings
train_cfg = dict(
rcnn=dict(
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.6,
neg_iou_thr=0.6,
min_pos_iou=0.6,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.25,
neg_pos_ub=-1,
add_gt_as_proposals=True),
pos_weight=-1,
debug=False))
test_cfg = dict(
rcnn=dict(
score_thr=1e-3, nms=dict(type='nms', iou_thr=0.5), max_per_img=100))
# dataset settings
dataset_type = 'CocoDataset'
data_root = 'data/coco/'
img_norm_cfg = dict(
mean=[102.9801, 115.9465, 122.7717], std=[1.0, 1.0, 1.0], to_rgb=False)
data = dict(
imgs_per_gpu=2,
workers_per_gpu=2,
train=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_train2017.json',
img_prefix=data_root + 'train2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
num_max_proposals=300,
proposal_file=data_root + 'proposals/ga_rpn_r50_fpn_1x_train2017.pkl',
flip_ratio=0.5,
with_mask=False,
with_crowd=True,
with_label=True),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
num_max_proposals=300,
proposal_file=data_root + 'proposals/ga_rpn_r50_fpn_1x_val2017.pkl',
flip_ratio=0,
with_mask=False,
with_crowd=True,
with_label=True),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
num_max_proposals=300,
proposal_file=data_root + 'proposals/ga_rpn_r50_fpn_1x_val2017.pkl',
flip_ratio=0,
with_mask=False,
with_label=False,
test_mode=True))
# optimizer
optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[8, 11])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
interval=50,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook')
])
# yapf:enable
# runtime settings
total_epochs = 12
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/ga_fast_rcnn_r50_caffe_fpn_1x'
load_from = None
resume_from = None
workflow = [('train', 1)]
configs/guided_anchoring/ga_faster_r50_caffe_fpn_1x.py 0 → 100644
View file @ 89022a26
# model settings
model = dict(
type='FasterRCNN',
pretrained='open-mmlab://resnet50_caffe',
backbone=dict(
type='ResNet',
depth=50,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
norm_cfg=dict(type='BN', requires_grad=False),
norm_eval=True,
style='caffe'),
neck=dict(
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=256,
num_outs=5),
rpn_head=dict(
type='GARPNHead',
in_channels=256,
feat_channels=256,
octave_base_scale=8,
scales_per_octave=3,
octave_ratios=[0.5, 1.0, 2.0],
anchor_strides=[4, 8, 16, 32, 64],
anchor_base_sizes=None,
anchoring_means=[.0, .0, .0, .0],
anchoring_stds=[0.07, 0.07, 0.14, 0.14],
target_means=(.0, .0, .0, .0),
target_stds=[0.07, 0.07, 0.11, 0.11],
loc_filter_thr=0.01,
loss_loc=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_shape=dict(
type='IoULoss', style='bounded', beta=0.2, loss_weight=1.0),
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)),
bbox_roi_extractor=dict(
type='SingleRoIExtractor',
roi_layer=dict(type='RoIAlign', out_size=7, sample_num=2),
out_channels=256,
featmap_strides=[4, 8, 16, 32]),
bbox_head=dict(
type='SharedFCBBoxHead',
num_fcs=2,
in_channels=256,
fc_out_channels=1024,
roi_feat_size=7,
num_classes=81,
target_means=[0., 0., 0., 0.],
target_stds=[0.05, 0.05, 0.1, 0.1],
reg_class_agnostic=False,
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))
# model training and testing settings
train_cfg = dict(
rpn=dict(
ga_assigner=dict(
type='ApproxMaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.3,
min_pos_iou=0.3,
ignore_iof_thr=-1),
ga_sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.3,
min_pos_iou=0.3,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
allowed_border=-1,
pos_weight=-1,
center_ratio=0.2,
ignore_ratio=0.5,
debug=False),
rpn_proposal=dict(
nms_across_levels=False,
nms_pre=2000,
nms_post=2000,
max_num=300,
nms_thr=0.7,
min_bbox_size=0),
rcnn=dict(
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.6,
neg_iou_thr=0.6,
min_pos_iou=0.6,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.25,
neg_pos_ub=-1,
add_gt_as_proposals=True),
pos_weight=-1,
debug=False))
test_cfg = dict(
rpn=dict(
nms_across_levels=False,
nms_pre=1000,
nms_post=1000,
max_num=300,
nms_thr=0.7,
min_bbox_size=0),
rcnn=dict(
score_thr=1e-3, nms=dict(type='nms', iou_thr=0.5), max_per_img=100))
# dataset settings
dataset_type = 'CocoDataset'
data_root = 'data/coco/'
img_norm_cfg = dict(
mean=[102.9801, 115.9465, 122.7717], std=[1.0, 1.0, 1.0], to_rgb=False)
data = dict(
imgs_per_gpu=2,
workers_per_gpu=2,
train=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_train2017.json',
img_prefix=data_root + 'train2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0.5,
with_mask=False,
with_crowd=True,
with_label=True),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_crowd=True,
with_label=True),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_label=False,
test_mode=True))
# optimizer
optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[8, 11])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
interval=50,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook')
])
# yapf:enable
# runtime settings
total_epochs = 12
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/ga_faster_rcnn_r50_caffe_fpn_1x'
load_from = None
resume_from = None
workflow = [('train', 1)]
configs/guided_anchoring/ga_faster_x101_32x4d_fpn_1x.py 0 → 100644
View file @ 89022a26
# model settings
model = dict(
type='FasterRCNN',
pretrained='open-mmlab://resnext101_32x4d',
backbone=dict(
type='ResNeXt',
depth=101,
groups=32,
base_width=4,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
style='pytorch'),
neck=dict(
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=256,
num_outs=5),
rpn_head=dict(
type='GARPNHead',
in_channels=256,
feat_channels=256,
octave_base_scale=8,
scales_per_octave=3,
octave_ratios=[0.5, 1.0, 2.0],
anchor_strides=[4, 8, 16, 32, 64],
anchor_base_sizes=None,
anchoring_means=[.0, .0, .0, .0],
anchoring_stds=[0.07, 0.07, 0.14, 0.14],
target_means=(.0, .0, .0, .0),
target_stds=[0.07, 0.07, 0.11, 0.11],
loc_filter_thr=0.01,
loss_loc=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_shape=dict(
type='IoULoss', style='bounded', beta=0.2, loss_weight=1.0),
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)),
bbox_roi_extractor=dict(
type='SingleRoIExtractor',
roi_layer=dict(type='RoIAlign', out_size=7, sample_num=2),
out_channels=256,
featmap_strides=[4, 8, 16, 32]),
bbox_head=dict(
type='SharedFCBBoxHead',
num_fcs=2,
in_channels=256,
fc_out_channels=1024,
roi_feat_size=7,
num_classes=81,
target_means=[0., 0., 0., 0.],
target_stds=[0.05, 0.05, 0.1, 0.1],
reg_class_agnostic=False,
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))
# model training and testing settings
train_cfg = dict(
rpn=dict(
ga_assigner=dict(
type='ApproxMaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.3,
min_pos_iou=0.3,
ignore_iof_thr=-1),
ga_sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.3,
min_pos_iou=0.3,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
allowed_border=-1,
pos_weight=-1,
center_ratio=0.2,
ignore_ratio=0.5,
debug=False),
rpn_proposal=dict(
nms_across_levels=False,
nms_pre=2000,
nms_post=2000,
max_num=300,
nms_thr=0.7,
min_bbox_size=0),
rcnn=dict(
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.6,
neg_iou_thr=0.6,
min_pos_iou=0.6,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.25,
neg_pos_ub=-1,
add_gt_as_proposals=True),
pos_weight=-1,
debug=False))
test_cfg = dict(
rpn=dict(
nms_across_levels=False,
nms_pre=1000,
nms_post=1000,
max_num=300,
nms_thr=0.7,
min_bbox_size=0),
rcnn=dict(
score_thr=1e-3, nms=dict(type='nms', iou_thr=0.5), max_per_img=100))
# dataset settings
dataset_type = 'CocoDataset'
data_root = 'data/coco/'
img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
data = dict(
imgs_per_gpu=2,
workers_per_gpu=2,
train=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_train2017.json',
img_prefix=data_root + 'train2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0.5,
with_mask=False,
with_crowd=True,
with_label=True),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_crowd=True,
with_label=True),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_label=False,
test_mode=True))
# optimizer
optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[8, 11])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
interval=50,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook')
])
# yapf:enable
# runtime settings
total_epochs = 12
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/ga_faster_rcnn_x101_32x4d_fpn_1x'
load_from = None
resume_from = None
workflow = [('train', 1)]
configs/guided_anchoring/ga_retinanet_r50_caffe_fpn_1x.py 0 → 100644
View file @ 89022a26
# model settings
model = dict(
type='RetinaNet',
pretrained='open-mmlab://resnet50_caffe',
backbone=dict(
type='ResNet',
depth=50,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
norm_cfg=dict(type='BN', requires_grad=False),
norm_eval=True,
style='caffe'),
neck=dict(
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=256,
start_level=1,
add_extra_convs=True,
num_outs=5),
bbox_head=dict(
type='GARetinaHead',
num_classes=81,
in_channels=256,
stacked_convs=4,
feat_channels=256,
octave_base_scale=4,
scales_per_octave=3,
octave_ratios=[0.5, 1.0, 2.0],
anchor_strides=[8, 16, 32, 64, 128],
anchor_base_sizes=None,
anchoring_means=[.0, .0, .0, .0],
anchoring_stds=[1.0, 1.0, 1.0, 1.0],
target_means=(.0, .0, .0, .0),
target_stds=[1.0, 1.0, 1.0, 1.0],
loc_filter_thr=0.01,
loss_loc=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_shape=dict(
type='IoULoss', style='bounded', beta=0.2, loss_weight=1.0),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=0.04, loss_weight=1.0)))
# training and testing settings
train_cfg = dict(
ga_assigner=dict(
type='ApproxMaxIoUAssigner',
pos_iou_thr=0.5,
neg_iou_thr=0.4,
min_pos_iou=0.4,
ignore_iof_thr=-1),
ga_sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.5,
neg_iou_thr=0.5,
min_pos_iou=0.0,
ignore_iof_thr=-1),
allowed_border=-1,
pos_weight=-1,
center_ratio=0.2,
ignore_ratio=0.5,
debug=False)
test_cfg = dict(
nms_pre=1000,
min_bbox_size=0,
score_thr=0.05,
nms=dict(type='nms', iou_thr=0.5),
max_per_img=100)
# dataset settings
dataset_type = 'CocoDataset'
data_root = 'data/coco/'
img_norm_cfg = dict(
mean=[102.9801, 115.9465, 122.7717], std=[1.0, 1.0, 1.0], to_rgb=False)
data = dict(
imgs_per_gpu=2,
workers_per_gpu=2,
train=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_train2017.json',
img_prefix=data_root + 'train2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0.5,
with_mask=False,
with_crowd=False,
with_label=True),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_crowd=False,
with_label=True),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_crowd=False,
with_label=False,
test_mode=True))
# optimizer
optimizer = dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[8, 11])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
interval=50,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook')
])
# yapf:enable
# runtime settings
total_epochs = 12
device_ids = range(8)
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/ga_retinanet_r50_caffe_fpn_1x'
load_from = None
resume_from = None
workflow = [('train', 1)]
configs/guided_anchoring/ga_retinanet_x101_32x4d_fpn_1x.py 0 → 100644
View file @ 89022a26
# model settings
model = dict(
type='RetinaNet',
pretrained='open-mmlab://resnext101_32x4d',
backbone=dict(
type='ResNeXt',
depth=101,
groups=32,
base_width=4,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
style='pytorch'),
neck=dict(
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=256,
start_level=1,
add_extra_convs=True,
num_outs=5),
bbox_head=dict(
type='GARetinaHead',
num_classes=81,
in_channels=256,
stacked_convs=4,
feat_channels=256,
octave_base_scale=4,
scales_per_octave=3,
octave_ratios=[0.5, 1.0, 2.0],
anchor_strides=[8, 16, 32, 64, 128],
anchor_base_sizes=None,
anchoring_means=[.0, .0, .0, .0],
anchoring_stds=[1.0, 1.0, 1.0, 1.0],
target_means=(.0, .0, .0, .0),
target_stds=[1.0, 1.0, 1.0, 1.0],
loc_filter_thr=0.01,
loss_loc=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_shape=dict(
type='IoULoss', style='bounded', beta=0.2, loss_weight=1.0),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=0.04, loss_weight=1.0)))
# training and testing settings
train_cfg = dict(
ga_assigner=dict(
type='ApproxMaxIoUAssigner',
pos_iou_thr=0.5,
neg_iou_thr=0.4,
min_pos_iou=0.4,
ignore_iof_thr=-1),
ga_sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.5,
neg_iou_thr=0.5,
min_pos_iou=0.0,
ignore_iof_thr=-1),
allowed_border=-1,
pos_weight=-1,
center_ratio=0.2,
ignore_ratio=0.5,
debug=False)
test_cfg = dict(
nms_pre=1000,
min_bbox_size=0,
score_thr=0.05,
nms=dict(type='nms', iou_thr=0.5),
max_per_img=100)
# dataset settings
dataset_type = 'CocoDataset'
data_root = 'data/coco/'
img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
data = dict(
imgs_per_gpu=2,
workers_per_gpu=2,
train=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_train2017.json',
img_prefix=data_root + 'train2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0.5,
with_mask=False,
with_crowd=False,
with_label=True),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_crowd=False,
with_label=True),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_crowd=False,
with_label=False,
test_mode=True))
# optimizer
optimizer = dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[8, 11])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
interval=50,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook')
])
# yapf:enable
# runtime settings
total_epochs = 12
device_ids = range(8)
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/ga_retinanet_x101_32x4d_fpn_1x'
load_from = None
resume_from = None
workflow = [('train', 1)]
configs/guided_anchoring/ga_rpn_r101_caffe_rpn_1x.py 0 → 100644
View file @ 89022a26
# model settings
model = dict(
type='RPN',
pretrained='open-mmlab://resnet101_caffe',
backbone=dict(
type='ResNet',
depth=101,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
norm_cfg=dict(type='BN', requires_grad=False),
norm_eval=True,
style='caffe'),
neck=dict(
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=256,
num_outs=5),
rpn_head=dict(
type='GARPNHead',
in_channels=256,
feat_channels=256,
octave_base_scale=8,
scales_per_octave=3,
octave_ratios=[0.5, 1.0, 2.0],
anchor_strides=[4, 8, 16, 32, 64],
anchor_base_sizes=None,
anchoring_means=[.0, .0, .0, .0],
anchoring_stds=[0.07, 0.07, 0.14, 0.14],
target_means=(.0, .0, .0, .0),
target_stds=[0.07, 0.07, 0.11, 0.11],
loc_filter_thr=0.01,
loss_loc=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_shape=dict(
type='IoULoss', style='bounded', beta=0.2, loss_weight=1.0),
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))
# model training and testing settings
train_cfg = dict(
rpn=dict(
ga_assigner=dict(
type='ApproxMaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.3,
min_pos_iou=0.3,
ignore_iof_thr=-1),
ga_sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.3,
min_pos_iou=0.3,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
allowed_border=-1,
pos_weight=-1,
center_ratio=0.2,
ignore_ratio=0.5,
debug=False))
test_cfg = dict(
rpn=dict(
nms_across_levels=False,
nms_pre=2000,
nms_post=2000,
max_num=2000,
nms_thr=0.7,
min_bbox_size=0))
# dataset settings
dataset_type = 'CocoDataset'
data_root = 'data/coco/'
img_norm_cfg = dict(
mean=[102.9801, 115.9465, 122.7717], std=[1.0, 1.0, 1.0], to_rgb=False)
data = dict(
imgs_per_gpu=2,
workers_per_gpu=2,
train=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_train2017.json',
img_prefix=data_root + 'train2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0.5,
with_mask=False,
with_crowd=False,
with_label=False),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_crowd=False,
with_label=False),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_label=False,
test_mode=True))
# optimizer
optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001)
# runner configs
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[8, 11])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
interval=50,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook')
])
# yapf:enable
# runtime settings
total_epochs = 12
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/ga_rpn_r101_caffe_fpn_1x'
load_from = None
resume_from = None
workflow = [('train', 1)]
configs/guided_anchoring/ga_rpn_r50_caffe_fpn_1x.py 0 → 100644
View file @ 89022a26
# model settings
model = dict(
type='RPN',
pretrained='open-mmlab://resnet50_caffe',
backbone=dict(
type='ResNet',
depth=50,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
norm_cfg=dict(type='BN', requires_grad=False),
norm_eval=True,
style='caffe'),
neck=dict(
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=256,
num_outs=5),
rpn_head=dict(
type='GARPNHead',
in_channels=256,
feat_channels=256,
octave_base_scale=8,
scales_per_octave=3,
octave_ratios=[0.5, 1.0, 2.0],
anchor_strides=[4, 8, 16, 32, 64],
anchor_base_sizes=None,
anchoring_means=[.0, .0, .0, .0],
anchoring_stds=[0.07, 0.07, 0.14, 0.14],
target_means=(.0, .0, .0, .0),
target_stds=[0.07, 0.07, 0.11, 0.11],
loc_filter_thr=0.01,
loss_loc=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_shape=dict(
type='IoULoss', style='bounded', beta=0.2, loss_weight=1.0),
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))
# model training and testing settings
train_cfg = dict(
rpn=dict(
ga_assigner=dict(
type='ApproxMaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.3,
min_pos_iou=0.3,
ignore_iof_thr=-1),
ga_sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.3,
min_pos_iou=0.3,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
allowed_border=-1,
pos_weight=-1,
center_ratio=0.2,
ignore_ratio=0.5,
debug=False))
test_cfg = dict(
rpn=dict(
nms_across_levels=False,
nms_pre=2000,
nms_post=2000,
max_num=2000,
nms_thr=0.7,
min_bbox_size=0))
# dataset settings
dataset_type = 'CocoDataset'
data_root = 'data/coco/'
img_norm_cfg = dict(
mean=[102.9801, 115.9465, 122.7717], std=[1.0, 1.0, 1.0], to_rgb=False)
data = dict(
imgs_per_gpu=2,
workers_per_gpu=2,
train=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_train2017.json',
img_prefix=data_root + 'train2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0.5,
with_mask=False,
with_crowd=False,
with_label=False),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_crowd=False,
with_label=False),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_label=False,
test_mode=True))
# optimizer
optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001)
# runner configs
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[8, 11])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
interval=50,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook')
])
# yapf:enable
# runtime settings
total_epochs = 12
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/ga_rpn_r50_caffe_fpn_1x'
load_from = None
resume_from = None
workflow = [('train', 1)]
configs/guided_anchoring/ga_rpn_x101_32x4d_fpn_1x.py 0 → 100644
View file @ 89022a26
# model settings
model = dict(
type='RPN',
pretrained='open-mmlab://resnext101_32x4d',
backbone=dict(
type='ResNeXt',
depth=101,
groups=32,
base_width=4,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
style='pytorch'),
neck=dict(
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=256,
num_outs=5),
rpn_head=dict(
type='GARPNHead',
in_channels=256,
feat_channels=256,
octave_base_scale=8,
scales_per_octave=3,
octave_ratios=[0.5, 1.0, 2.0],
anchor_strides=[4, 8, 16, 32, 64],
anchor_base_sizes=None,
anchoring_means=[.0, .0, .0, .0],
anchoring_stds=[0.07, 0.07, 0.14, 0.14],
target_means=(.0, .0, .0, .0),
target_stds=[0.07, 0.07, 0.11, 0.11],
loc_filter_thr=0.01,
loss_loc=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
loss_shape=dict(
type='IoULoss', style='bounded', beta=0.2, loss_weight=1.0),
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)))
# model training and testing settings
train_cfg = dict(
rpn=dict(
ga_assigner=dict(
type='ApproxMaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.3,
min_pos_iou=0.3,
ignore_iof_thr=-1),
ga_sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.3,
min_pos_iou=0.3,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=256,
pos_fraction=0.5,
neg_pos_ub=-1,
add_gt_as_proposals=False),
allowed_border=-1,
pos_weight=-1,
center_ratio=0.2,
ignore_ratio=0.5,
debug=False))
test_cfg = dict(
rpn=dict(
nms_across_levels=False,
nms_pre=2000,
nms_post=2000,
max_num=2000,
nms_thr=0.7,
min_bbox_size=0))
# dataset settings
dataset_type = 'CocoDataset'
data_root = 'data/coco/'
img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
data = dict(
imgs_per_gpu=2,
workers_per_gpu=2,
train=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_train2017.json',
img_prefix=data_root + 'train2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0.5,
with_mask=False,
with_crowd=False,
with_label=False),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_crowd=False,
with_label=False),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
img_scale=(1333, 800),
img_norm_cfg=img_norm_cfg,
size_divisor=32,
flip_ratio=0,
with_mask=False,
with_label=False,
test_mode=True))
# optimizer
optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0001)
# runner configs
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[8, 11])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
interval=50,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook')
])
# yapf:enable
# runtime settings
total_epochs = 12
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/ga_rpn_x101_32x4d_fpn_1x'
load_from = None
resume_from = None
workflow = [('train', 1)]
mmdet/core/anchor/__init__.py
View file @ 89022a26
from .anchor_generator import AnchorGenerator from .anchor_generator import AnchorGenerator
from .anchor_target import anchor_target from .anchor_target import anchor_target, anchor_inside_flags
from .guided_anchor_target import ga_loc_target, ga_shape_target
__all__ = ['AnchorGenerator', 'anchor_target'] __all__ = [
'AnchorGenerator', 'anchor_target', 'anchor_inside_flags', 'ga_loc_target',
'ga_shape_target'
]
mmdet/core/anchor/guided_anchor_target.py 0 → 100644
View file @ 89022a26
import torch
from ..bbox import build_assigner, build_sampler, PseudoSampler
from ..utils import unmap, multi_apply
def calc_region(bbox, ratio, featmap_size=None):
"""Calculate a proportional bbox region.
The bbox center are fixed and the new h' and w' is h * ratio and w * ratio.
Args:
bbox (Tensor): Bboxes to calculate regions, shape (n, 4)
ratio (float): Ratio of the output region.
featmap_size (tuple): Feature map size used for clipping the boundary.
Returns:
tuple: x1, y1, x2, y2
"""
x1 = torch.round((1 - ratio) * bbox[0] + ratio * bbox[2]).long()
y1 = torch.round((1 - ratio) * bbox[1] + ratio * bbox[3]).long()
x2 = torch.round(ratio * bbox[0] + (1 - ratio) * bbox[2]).long()
y2 = torch.round(ratio * bbox[1] + (1 - ratio) * bbox[3]).long()
if featmap_size is not None:
x1 = x1.clamp(min=0, max=featmap_size[1] - 1)
y1 = y1.clamp(min=0, max=featmap_size[0] - 1)
x2 = x2.clamp(min=0, max=featmap_size[1] - 1)
y2 = y2.clamp(min=0, max=featmap_size[0] - 1)
return (x1, y1, x2, y2)
def ga_loc_target(gt_bboxes_list,
featmap_sizes,
anchor_scale,
anchor_strides,
center_ratio=0.2,
ignore_ratio=0.5):
"""Compute location targets for guided anchoring.
Each feature map is divided into positive, negative and ignore regions.
- positive regions: target 1, weight 1
- ignore regions: target 0, weight 0
- negative regions: target 0, weight 0.1
Args:
gt_bboxes_list (list[Tensor]): Gt bboxes of each image.
featmap_sizes (list[tuple]): Multi level sizes of each feature maps.
anchor_scale (int): Anchor scale.
anchor_strides ([list[int]]): Multi level anchor strides.
center_ratio (float): Ratio of center region.
ignore_ratio (float): Ratio of ignore region.
Returns:
tuple
"""
img_per_gpu = len(gt_bboxes_list)
num_lvls = len(featmap_sizes)
r1 = (1 - center_ratio) / 2
r2 = (1 - ignore_ratio) / 2
all_loc_targets = []
all_loc_weights = []
all_ignore_map = []
for lvl_id in range(num_lvls):
h, w = featmap_sizes[lvl_id]
loc_targets = torch.zeros(img_per_gpu,
1,
h,
w,
device=gt_bboxes_list[0].device,
dtype=torch.float32)
loc_weights = torch.full_like(loc_targets, -1)
ignore_map = torch.zeros_like(loc_targets)
all_loc_targets.append(loc_targets)
all_loc_weights.append(loc_weights)
all_ignore_map.append(ignore_map)
for img_id in range(img_per_gpu):
gt_bboxes = gt_bboxes_list[img_id]
scale = torch.sqrt((gt_bboxes[:, 2] - gt_bboxes[:, 0] + 1) *
(gt_bboxes[:, 3] - gt_bboxes[:, 1] + 1))
min_anchor_size = scale.new_full(
(1, ), float(anchor_scale * anchor_strides[0]))
# assign gt bboxes to different feature levels w.r.t. their scales
target_lvls = torch.floor(
torch.log2(scale) - torch.log2(min_anchor_size) + 0.5)
target_lvls = target_lvls.clamp(min=0, max=num_lvls - 1).long()
for gt_id in range(gt_bboxes.size(0)):
lvl = target_lvls[gt_id].item()
# rescaled to corresponding feature map
gt_ = gt_bboxes[gt_id, :4] / anchor_strides[lvl]
# calculate ignore regions
ignore_x1, ignore_y1, ignore_x2, ignore_y2 = calc_region(
gt_, r2, featmap_sizes[lvl])
# calculate positive (center) regions
ctr_x1, ctr_y1, ctr_x2, ctr_y2 = calc_region(
gt_, r1, featmap_sizes[lvl])
all_loc_targets[lvl][img_id, 0, ctr_y1:ctr_y2 + 1, ctr_x1:ctr_x2 +
1] = 1
all_loc_weights[lvl][img_id, 0, ignore_y1:ignore_y2 +
1, ignore_x1:ignore_x2 + 1] = 0
all_loc_weights[lvl][img_id, 0, ctr_y1:ctr_y2 + 1, ctr_x1:ctr_x2 +
1] = 1
# calculate ignore map on nearby low level feature
if lvl > 0:
d_lvl = lvl - 1
# rescaled to corresponding feature map
gt_ = gt_bboxes[gt_id, :4] / anchor_strides[d_lvl]
ignore_x1, ignore_y1, ignore_x2, ignore_y2 = calc_region(
gt_, r2, featmap_sizes[d_lvl])
all_ignore_map[d_lvl][img_id, 0, ignore_y1:ignore_y2 +
1, ignore_x1:ignore_x2 + 1] = 1
# calculate ignore map on nearby high level feature
if lvl < num_lvls - 1:
u_lvl = lvl + 1
# rescaled to corresponding feature map
gt_ = gt_bboxes[gt_id, :4] / anchor_strides[u_lvl]
ignore_x1, ignore_y1, ignore_x2, ignore_y2 = calc_region(
gt_, r2, featmap_sizes[u_lvl])
all_ignore_map[u_lvl][img_id, 0, ignore_y1:ignore_y2 +
1, ignore_x1:ignore_x2 + 1] = 1
for lvl_id in range(num_lvls):
# ignore negative regions w.r.t. ignore map
all_loc_weights[lvl_id][(all_loc_weights[lvl_id] < 0)
& (all_ignore_map[lvl_id] > 0)] = 0
# set negative regions with weight 0.1
all_loc_weights[lvl_id][all_loc_weights[lvl_id] < 0] = 0.1
# loc average factor to balance loss
loc_avg_factor = sum(
[t.size(0) * t.size(-1) * t.size(-2) for t in all_loc_targets]) / 200
return all_loc_targets, all_loc_weights, loc_avg_factor
def ga_shape_target(approx_list,
inside_flag_list,
square_list,
gt_bboxes_list,
img_metas,
approxs_per_octave,
cfg,
gt_bboxes_ignore_list=None,
sampling=True,
unmap_outputs=True):
"""Compute guided anchoring targets.
Args:
approx_list (list[list]): Multi level approxs of each image.
inside_flag_list (list[list]): Multi level inside flags of each image.
square_list (list[list]): Multi level squares of each image.
gt_bboxes_list (list[Tensor]): Ground truth bboxes of each image.
img_metas (list[dict]): Meta info of each image.
approxs_per_octave (int): number of approxs per octave
cfg (dict): RPN train configs.
gt_bboxes_ignore_list (list[Tensor]): ignore list of gt bboxes.
sampling (bool): sampling or not.
unmap_outputs (bool): unmap outputs or not.
Returns:
tuple
"""
num_imgs = len(img_metas)
assert len(approx_list) == len(inside_flag_list) == len(
square_list) == num_imgs
# anchor number of multi levels
num_level_squares = [squares.size(0) for squares in square_list[0]]
# concat all level anchors and flags to a single tensor
inside_flag_flat_list = []
approx_flat_list = []
square_flat_list = []
for i in range(num_imgs):
assert len(square_list[i]) == len(inside_flag_list[i])
inside_flag_flat_list.append(torch.cat(inside_flag_list[i]))
approx_flat_list.append(torch.cat(approx_list[i]))
square_flat_list.append(torch.cat(square_list[i]))
# compute targets for each image
if gt_bboxes_ignore_list is None:
gt_bboxes_ignore_list = [None for _ in range(num_imgs)]
(all_bbox_anchors, all_bbox_gts, all_bbox_weights, pos_inds_list,
neg_inds_list) = multi_apply(ga_shape_target_single,
approx_flat_list,
inside_flag_flat_list,
square_flat_list,
gt_bboxes_list,
gt_bboxes_ignore_list,
img_metas,
approxs_per_octave=approxs_per_octave,
cfg=cfg,
sampling=sampling,
unmap_outputs=unmap_outputs)
# no valid anchors
if any([bbox_anchors is None for bbox_anchors in all_bbox_anchors]):
return None
# sampled anchors of all images
num_total_pos = sum([max(inds.numel(), 1) for inds in pos_inds_list])
num_total_neg = sum([max(inds.numel(), 1) for inds in neg_inds_list])
# split targets to a list w.r.t. multiple levels
bbox_anchors_list = images_to_levels(all_bbox_anchors, num_level_squares)
bbox_gts_list = images_to_levels(all_bbox_gts, num_level_squares)
bbox_weights_list = images_to_levels(all_bbox_weights, num_level_squares)
return (bbox_anchors_list, bbox_gts_list, bbox_weights_list, num_total_pos,
num_total_neg)
def images_to_levels(target, num_level_anchors):
"""Convert targets by image to targets by feature level.
[target_img0, target_img1] -> [target_level0, target_level1, ...]
"""
target = torch.stack(target, 0)
level_targets = []
start = 0
for n in num_level_anchors:
end = start + n
level_targets.append(target[:, start:end].squeeze(0))
start = end
return level_targets
def ga_shape_target_single(flat_approxs,
inside_flags,
flat_squares,
gt_bboxes,
gt_bboxes_ignore,
img_meta,
approxs_per_octave,
cfg,
sampling=True,
unmap_outputs=True):
"""Compute guided anchoring targets.
This function returns sampled anchors and gt bboxes directly
rather than calculates regression targets.
Args:
flat_approxs (Tensor): flat approxs of a single image,
shape (n, 4)
inside_flags (Tensor): inside flags of a single image,
shape (n, ).
flat_squares (Tensor): flat squares of a single image,
shape (approxs_per_octave * n, 4)
gt_bboxes (Tensor): Ground truth bboxes of a single image.
img_meta (dict): Meta info of a single image.
approxs_per_octave (int): number of approxs per octave
cfg (dict): RPN train configs.
sampling (bool): sampling or not.
unmap_outputs (bool): unmap outputs or not.
Returns:
tuple
"""
if not inside_flags.any():
return (None, ) * 6
# assign gt and sample anchors
expand_inside_flags = inside_flags[:, None].expand(
-1, approxs_per_octave).reshape(-1)
approxs = flat_approxs[expand_inside_flags, :]
squares = flat_squares[inside_flags, :]
bbox_assigner = build_assigner(cfg.ga_assigner)
assign_result = bbox_assigner.assign(approxs, squares, approxs_per_octave,
gt_bboxes, gt_bboxes_ignore)
if sampling:
bbox_sampler = build_sampler(cfg.ga_sampler)
else:
bbox_sampler = PseudoSampler()
sampling_result = bbox_sampler.sample(assign_result, squares, gt_bboxes)
bbox_anchors = torch.zeros_like(squares)
bbox_gts = torch.zeros_like(squares)
bbox_weights = torch.zeros_like(squares)
pos_inds = sampling_result.pos_inds
neg_inds = sampling_result.neg_inds
if len(pos_inds) > 0:
bbox_anchors[pos_inds, :] = sampling_result.pos_bboxes
bbox_gts[pos_inds, :] = sampling_result.pos_gt_bboxes
bbox_weights[pos_inds, :] = 1.0
# map up to original set of anchors
if unmap_outputs:
num_total_anchors = flat_squares.size(0)
bbox_anchors = unmap(bbox_anchors, num_total_anchors, inside_flags)
bbox_gts = unmap(bbox_gts, num_total_anchors, inside_flags)
bbox_weights = unmap(bbox_weights, num_total_anchors, inside_flags)
return (bbox_anchors, bbox_gts, bbox_weights, pos_inds, neg_inds)
mmdet/core/bbox/assigners/__init__.py
View file @ 89022a26
from .base_assigner import BaseAssigner from .base_assigner import BaseAssigner
from .max_iou_assigner import MaxIoUAssigner from .max_iou_assigner import MaxIoUAssigner
from .approx_max_iou_assigner import ApproxMaxIoUAssigner
from .assign_result import AssignResult from .assign_result import AssignResult
__all__ = ['BaseAssigner', 'MaxIoUAssigner', 'AssignResult'] __all__ = [
'BaseAssigner', 'MaxIoUAssigner', 'ApproxMaxIoUAssigner', 'AssignResult'
]
mmdet/core/bbox/assigners/approx_max_iou_assigner.py 0 → 100644
View file @ 89022a26
import torch
from .max_iou_assigner import MaxIoUAssigner
from ..geometry import bbox_overlaps
class ApproxMaxIoUAssigner(MaxIoUAssigner):
"""Assign a corresponding gt bbox or background to each bbox.
Each proposals will be assigned with `-1`, `0`, or a positive integer
indicating the ground truth index.
- -1: don't care
- 0: negative sample, no assigned gt
- positive integer: positive sample, index (1-based) of assigned gt
Args:
pos_iou_thr (float): IoU threshold for positive bboxes.
neg_iou_thr (float or tuple): IoU threshold for negative bboxes.
min_pos_iou (float): Minimum iou for a bbox to be considered as a
positive bbox. Positive samples can have smaller IoU than
pos_iou_thr due to the 4th step (assign max IoU sample to each gt).
gt_max_assign_all (bool): Whether to assign all bboxes with the same
highest overlap with some gt to that gt.
ignore_iof_thr (float): IoF threshold for ignoring bboxes (if
`gt_bboxes_ignore` is specified). Negative values mean not
ignoring any bboxes.
ignore_wrt_candidates (bool): Whether to compute the iof between
`bboxes` and `gt_bboxes_ignore`, or the contrary.
"""
def __init__(self,
pos_iou_thr,
neg_iou_thr,
min_pos_iou=.0,
gt_max_assign_all=True,
ignore_iof_thr=-1,
ignore_wrt_candidates=True):
self.pos_iou_thr = pos_iou_thr
self.neg_iou_thr = neg_iou_thr
self.min_pos_iou = min_pos_iou
self.gt_max_assign_all = gt_max_assign_all
self.ignore_iof_thr = ignore_iof_thr
self.ignore_wrt_candidates = ignore_wrt_candidates
def assign(self,
approxs,
squares,
approxs_per_octave,
gt_bboxes,
gt_bboxes_ignore=None,
gt_labels=None):
"""Assign gt to approxs.
This method assign a gt bbox to each group of approxs (bboxes),
each group of approxs is represent by a base approx (bbox) and
will be assigned with -1, 0, or a positive number.
-1 means don't care, 0 means negative sample,
positive number is the index (1-based) of assigned gt.
The assignment is done in following steps, the order matters.
1. assign every bbox to -1
2. use the max IoU of each group of approxs to assign
2. assign proposals whose iou with all gts < neg_iou_thr to 0
3. for each bbox, if the iou with its nearest gt >= pos_iou_thr,
assign it to that bbox
4. for each gt bbox, assign its nearest proposals (may be more than
one) to itself
Args:
approxs (Tensor): Bounding boxes to be assigned,
shape(approxs_per_octave*n, 4).
squares (Tensor): Base Bounding boxes to be assigned,
shape(n, 4).
approxs_per_octave (int): number of approxs per octave
gt_bboxes (Tensor): Groundtruth boxes, shape (k, 4).
gt_bboxes_ignore (Tensor, optional): Ground truth bboxes that are
labelled as `ignored`, e.g., crowd boxes in COCO.
gt_labels (Tensor, optional): Label of gt_bboxes, shape (k, ).
Returns:
:obj:`AssignResult`: The assign result.
"""
if squares.shape[0] == 0 or gt_bboxes.shape[0] == 0:
raise ValueError('No gt or approxs')
num_squares = squares.size(0)
num_gts = gt_bboxes.size(0)
# re-organize anchors by approxs_per_octave x num_squares
approxs = torch.transpose(
approxs.view(num_squares, approxs_per_octave, 4), 0,
1).contiguous().view(-1, 4)
all_overlaps = bbox_overlaps(approxs, gt_bboxes)
overlaps, _ = all_overlaps.view(approxs_per_octave, num_squares,
num_gts).max(dim=0)
overlaps = torch.transpose(overlaps, 0, 1)
bboxes = squares[:, :4]
if (self.ignore_iof_thr > 0) and (gt_bboxes_ignore is not None) and (
gt_bboxes_ignore.numel() > 0):
if self.ignore_wrt_candidates:
ignore_overlaps = bbox_overlaps(bboxes,
gt_bboxes_ignore,
mode='iof')
ignore_max_overlaps, _ = ignore_overlaps.max(dim=1)
else:
ignore_overlaps = bbox_overlaps(gt_bboxes_ignore,
bboxes,
mode='iof')
ignore_max_overlaps, _ = ignore_overlaps.max(dim=0)
overlaps[:, ignore_max_overlaps > self.ignore_iof_thr] = -1
assign_result = self.assign_wrt_overlaps(overlaps, gt_labels)
return assign_result
mmdet/core/loss/__init__.py
View file @ 89022a26
from .losses import ( from .losses import (weighted_nll_loss, weighted_cross_entropy,
weighted_nll_loss, weighted_cross_entropy, weighted_binary_cross_entropy, weighted_binary_cross_entropy, sigmoid_focal_loss,
sigmoid_focal_loss, py_sigmoid_focal_loss, weighted_sigmoid_focal_loss, py_sigmoid_focal_loss, weighted_sigmoid_focal_loss,
mask_cross_entropy, smooth_l1_loss, weighted_smoothl1, accuracy, iou_loss) mask_cross_entropy, smooth_l1_loss, weighted_smoothl1,
bounded_iou_loss, weighted_iou_loss, iou_loss, accuracy)
__all__ = [ __all__ = [
'weighted_nll_loss', 'weighted_cross_entropy', 'weighted_nll_loss', 'weighted_cross_entropy',
'weighted_binary_cross_entropy', 'sigmoid_focal_loss', 'weighted_binary_cross_entropy', 'sigmoid_focal_loss',
'py_sigmoid_focal_loss', 'weighted_sigmoid_focal_loss', 'py_sigmoid_focal_loss', 'weighted_sigmoid_focal_loss',
'mask_cross_entropy', 'smooth_l1_loss', 'weighted_smoothl1', 'accuracy', 'mask_cross_entropy', 'smooth_l1_loss', 'weighted_smoothl1',
'iou_loss' 'bounded_iou_loss', 'weighted_iou_loss', 'iou_loss', 'accuracy'
] ]
mmdet/core/loss/losses.py
View file @ 89022a26
...@@ -44,8 +44,8 @@ def py_sigmoid_focal_loss(pred, ...@@ -44,8 +44,8 @@ def py_sigmoid_focal_loss(pred,
pt = (1 - pred_sigmoid) * target + pred_sigmoid * (1 - target) pt = (1 - pred_sigmoid) * target + pred_sigmoid * (1 - target)
weight = (alpha * target + (1 - alpha) * (1 - target)) * weight weight = (alpha * target + (1 - alpha) * (1 - target)) * weight
weight = weight * pt.pow(gamma) weight = weight * pt.pow(gamma)
loss = F.binary_cross_entropy_with_logits( loss = F.binary_cross_entropy_with_logits(pred, target,
pred, target, reduction='none') * weight reduction='none') * weight
reduction_enum = F._Reduction.get_enum(reduction) reduction_enum = F._Reduction.get_enum(reduction)
# none: 0, mean:1, sum: 2 # none: 0, mean:1, sum: 2
if reduction_enum == 0: if reduction_enum == 0:
...@@ -66,16 +66,17 @@ def weighted_sigmoid_focal_loss(pred, ...@@ -66,16 +66,17 @@ def weighted_sigmoid_focal_loss(pred,
if avg_factor is None: if avg_factor is None:
avg_factor = torch.sum(weight > 0).float().item() / num_classes + 1e-6 avg_factor = torch.sum(weight > 0).float().item() / num_classes + 1e-6
return torch.sum( return torch.sum(
sigmoid_focal_loss(pred, target, gamma, alpha, 'none') * weight.view( sigmoid_focal_loss(pred, target, gamma, alpha, 'none') *
-1, 1))[None] / avg_factor weight.view(-1, 1))[None] / avg_factor
def mask_cross_entropy(pred, target, label): def mask_cross_entropy(pred, target, label):
num_rois = pred.size()[0] num_rois = pred.size()[0]
inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device) inds = torch.arange(0, num_rois, dtype=torch.long, device=pred.device)
pred_slice = pred[inds, label].squeeze(1) pred_slice = pred[inds, label].squeeze(1)
return F.binary_cross_entropy_with_logits( return F.binary_cross_entropy_with_logits(pred_slice,
pred_slice, target, reduction='mean')[None] target,
reduction='mean')[None]
def smooth_l1_loss(pred, target, beta=1.0, reduction='mean'): def smooth_l1_loss(pred, target, beta=1.0, reduction='mean'):
...@@ -101,6 +102,85 @@ def weighted_smoothl1(pred, target, weight, beta=1.0, avg_factor=None): ...@@ -101,6 +102,85 @@ def weighted_smoothl1(pred, target, weight, beta=1.0, avg_factor=None):
return torch.sum(loss * weight)[None] / avg_factor return torch.sum(loss * weight)[None] / avg_factor
def bounded_iou_loss(pred, target, beta=0.2, eps=1e-3, reduction='mean'):
"""Improving Object Localization with Fitness NMS and Bounded IoU Loss,
https://arxiv.org/abs/1711.00164.
Args:
pred (tensor): Predicted bboxes.
target (tensor): Target bboxes.
beta (float): beta parameter in smoothl1.
eps (float): eps to avoid NaN.
reduction (str): Reduction type.
"""
pred_ctrx = (pred[:, 0] + pred[:, 2]) * 0.5
pred_ctry = (pred[:, 1] + pred[:, 3]) * 0.5
pred_w = pred[:, 2] - pred[:, 0] + 1
pred_h = pred[:, 3] - pred[:, 1] + 1
with torch.no_grad():
target_ctrx = (target[:, 0] + target[:, 2]) * 0.5
target_ctry = (target[:, 1] + target[:, 3]) * 0.5
target_w = target[:, 2] - target[:, 0] + 1
target_h = target[:, 3] - target[:, 1] + 1
dx = target_ctrx - pred_ctrx
dy = target_ctry - pred_ctry
loss_dx = 1 - torch.max(
(target_w - 2 * dx.abs()) /
(target_w + 2 * dx.abs() + eps), torch.zeros_like(dx))
loss_dy = 1 - torch.max(
(target_h - 2 * dy.abs()) /
(target_h + 2 * dy.abs() + eps), torch.zeros_like(dy))
loss_dw = 1 - torch.min(target_w / (pred_w + eps), pred_w /
(target_w + eps))
loss_dh = 1 - torch.min(target_h / (pred_h + eps), pred_h /
(target_h + eps))
loss_comb = torch.stack([loss_dx, loss_dy, loss_dw, loss_dh],
dim=-1).view(loss_dx.size(0), -1)
loss = torch.where(loss_comb < beta, 0.5 * loss_comb * loss_comb / beta,
loss_comb - 0.5 * beta)
reduction_enum = F._Reduction.get_enum(reduction)
# none: 0, mean:1, sum: 2
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.sum() / pred.numel()
elif reduction_enum == 2:
return loss.sum()
def weighted_iou_loss(pred,
target,
weight,
style='naive',
beta=0.2,
eps=1e-3,
avg_factor=None):
if style not in ['bounded', 'naive']:
raise ValueError('Only support bounded iou loss and naive iou loss.')
inds = torch.nonzero(weight[:, 0] > 0)
if avg_factor is None:
avg_factor = inds.numel() + 1e-6
if inds.numel() > 0:
inds = inds.squeeze(1)
else:
return (pred * weight).sum()[None] / avg_factor
if style == 'bounded':
loss = bounded_iou_loss(pred[inds],
target[inds],
beta=beta,
eps=eps,
reduction='sum')
else:
loss = iou_loss(pred[inds], target[inds], reduction='sum')
loss = loss[None] / avg_factor
return loss
def accuracy(pred, target, topk=1): def accuracy(pred, target, topk=1):
if isinstance(topk, int): if isinstance(topk, int):
topk = (topk, ) topk = (topk, )
...@@ -125,8 +205,9 @@ def _expand_binary_labels(labels, label_weights, label_channels): ...@@ -125,8 +205,9 @@ def _expand_binary_labels(labels, label_weights, label_channels):
inds = torch.nonzero(labels >= 1).squeeze() inds = torch.nonzero(labels >= 1).squeeze()
if inds.numel() > 0: if inds.numel() > 0:
bin_labels[inds, labels[inds] - 1] = 1 bin_labels[inds, labels[inds] - 1] = 1
bin_label_weights = label_weights.view(-1, 1).expand( bin_label_weights = label_weights.view(-1,
label_weights.size(0), label_channels) 1).expand(label_weights.size(0),
label_channels)
return bin_labels, bin_label_weights return bin_labels, bin_label_weights
......
mmdet/models/anchor_heads/__init__.py
View file @ 89022a26
from .anchor_head import AnchorHead from .anchor_head import AnchorHead
from .guided_anchor_head import GuidedAnchorHead, FeatureAdaption
from .fcos_head import FCOSHead from .fcos_head import FCOSHead
from .retina_head import RetinaHead
from .rpn_head import RPNHead from .rpn_head import RPNHead
from .ga_rpn_head import GARPNHead
from .retina_head import RetinaHead
from .ga_retina_head import GARetinaHead
from .ssd_head import SSDHead from .ssd_head import SSDHead
__all__ = ['AnchorHead', 'RPNHead', 'RetinaHead', 'SSDHead', 'FCOSHead'] __all__ = [
'AnchorHead', 'GuidedAnchorHead', 'FeatureAdaption', 'RPNHead',
'GARPNHead', 'RetinaHead', 'GARetinaHead', 'SSDHead', 'FCOSHead'
]
mmdet/models/anchor_heads/ga_retina_head.py 0 → 100644
View file @ 89022a26
import torch.nn as nn
from mmcv.cnn import normal_init
from .guided_anchor_head import GuidedAnchorHead, FeatureAdaption
from ..registry import HEADS
from ..utils import bias_init_with_prob, ConvModule
from mmdet.ops import MaskedConv2d
@HEADS.register_module
class GARetinaHead(GuidedAnchorHead):
"""Guided-Anchor-based RetinaNet head."""
def __init__(self,
num_classes,
in_channels,
stacked_convs=4,
conv_cfg=None,
norm_cfg=None,
**kwargs):
self.stacked_convs = stacked_convs
self.conv_cfg = conv_cfg
self.norm_cfg = norm_cfg
super(GARetinaHead, self).__init__(num_classes, in_channels, **kwargs)
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)
def init_weights(self):
for m in self.cls_convs:
normal_init(m.conv, std=0.01)
for m in self.reg_convs:
normal_init(m.conv, std=0.01)
self.feature_adaption_cls.init_weights()
self.feature_adaption_reg.init_weights()
bias_cls = bias_init_with_prob(0.01)
normal_init(self.conv_loc, std=0.01, bias=bias_cls)
normal_init(self.conv_shape, std=0.01)
normal_init(self.retina_cls, std=0.01, bias=bias_cls)
normal_init(self.retina_reg, std=0.01)
def forward_single(self, x):
cls_feat = x
reg_feat = x
for cls_conv in self.cls_convs:
cls_feat = cls_conv(cls_feat)
for reg_conv in self.reg_convs:
reg_feat = reg_conv(reg_feat)
loc_pred = self.conv_loc(cls_feat)
shape_pred = self.conv_shape(reg_feat)
cls_feat = self.feature_adaption_cls(cls_feat, shape_pred)
reg_feat = self.feature_adaption_reg(reg_feat, shape_pred)
if not self.training:
mask = loc_pred.sigmoid()[0] >= self.loc_filter_thr
else:
mask = None
cls_score = self.retina_cls(cls_feat, mask)
bbox_pred = self.retina_reg(reg_feat, mask)
return cls_score, bbox_pred, shape_pred, loc_pred
mmdet/models/anchor_heads/ga_rpn_head.py 0 → 100644
View file @ 89022a26
import torch
import torch.nn as nn
import torch.nn.functional as F
from mmcv.cnn import normal_init
from mmdet.core import delta2bbox
from mmdet.ops import nms
from .guided_anchor_head import GuidedAnchorHead
from ..registry import HEADS
@HEADS.register_module
class GARPNHead(GuidedAnchorHead):
"""Guided-Anchor-based RPN head."""
def __init__(self, in_channels, **kwargs):
super(GARPNHead, self).__init__(2, in_channels, **kwargs)
def _init_layers(self):
self.rpn_conv = nn.Conv2d(self.in_channels,
self.feat_channels,
3,
padding=1)
super(GARPNHead, self)._init_layers()
def init_weights(self):
normal_init(self.rpn_conv, std=0.01)
super(GARPNHead, self).init_weights()
def forward_single(self, x):
x = self.rpn_conv(x)
x = F.relu(x, inplace=True)
(cls_score, bbox_pred, shape_pred,
loc_pred) = super(GARPNHead, self).forward_single(x)
return cls_score, bbox_pred, shape_pred, loc_pred
def loss(self,
cls_scores,
bbox_preds,
shape_preds,
loc_preds,
gt_bboxes,
img_metas,
cfg,
gt_bboxes_ignore=None):
losses = super(GARPNHead, self).loss(cls_scores,
bbox_preds,
shape_preds,
loc_preds,
gt_bboxes,
None,
img_metas,
cfg,
gt_bboxes_ignore=gt_bboxes_ignore)
return dict(loss_rpn_cls=losses['loss_cls'],
loss_rpn_bbox=losses['loss_bbox'],
loss_anchor_shape=losses['loss_shape'],
loss_anchor_loc=losses['loss_loc'])
def get_bboxes_single(self,
cls_scores,
bbox_preds,
mlvl_anchors,
mlvl_masks,
img_shape,
scale_factor,
cfg,
rescale=False):
mlvl_proposals = []
for idx in range(len(cls_scores)):
rpn_cls_score = cls_scores[idx]
rpn_bbox_pred = bbox_preds[idx]
anchors = mlvl_anchors[idx]
mask = mlvl_masks[idx]
assert rpn_cls_score.size()[-2:] == rpn_bbox_pred.size()[-2:]
# if no location is kept, end.
if mask.sum() == 0:
continue
rpn_cls_score = rpn_cls_score.permute(1, 2, 0)
if self.use_sigmoid_cls:
rpn_cls_score = rpn_cls_score.reshape(-1)
scores = rpn_cls_score.sigmoid()
else:
rpn_cls_score = rpn_cls_score.reshape(-1, 2)
scores = rpn_cls_score.softmax(dim=1)[:, 1]
# filter scores, bbox_pred w.r.t. mask.
# anchors are filtered in get_anchors() beforehand.
scores = scores[mask]
rpn_bbox_pred = rpn_bbox_pred.permute(1, 2, 0).reshape(-1,
4)[mask, :]
if scores.dim() == 0:
rpn_bbox_pred = rpn_bbox_pred.unsqueeze(0)
anchors = anchors.unsqueeze(0)
scores = scores.unsqueeze(0)
# filter anchors, bbox_pred, scores w.r.t. scores
if cfg.nms_pre > 0 and scores.shape[0] > cfg.nms_pre:
_, topk_inds = scores.topk(cfg.nms_pre)
rpn_bbox_pred = rpn_bbox_pred[topk_inds, :]
anchors = anchors[topk_inds, :]
scores = scores[topk_inds]
# get proposals w.r.t. anchors and rpn_bbox_pred
proposals = delta2bbox(anchors, rpn_bbox_pred, self.target_means,
self.target_stds, img_shape)
# filter out too small bboxes
if cfg.min_bbox_size > 0:
w = proposals[:, 2] - proposals[:, 0] + 1
h = proposals[:, 3] - proposals[:, 1] + 1
valid_inds = torch.nonzero((w >= cfg.min_bbox_size) &
(h >= cfg.min_bbox_size)).squeeze()
proposals = proposals[valid_inds, :]
scores = scores[valid_inds]
proposals = torch.cat([proposals, scores.unsqueeze(-1)], dim=-1)
# NMS in current level
proposals, _ = nms(proposals, cfg.nms_thr)
proposals = proposals[:cfg.nms_post, :]
mlvl_proposals.append(proposals)
proposals = torch.cat(mlvl_proposals, 0)
if cfg.nms_across_levels:
# NMS across multi levels
proposals, _ = nms(proposals, cfg.nms_thr)
proposals = proposals[:cfg.max_num, :]
else:
scores = proposals[:, 4]
num = min(cfg.max_num, proposals.shape[0])
_, topk_inds = scores.topk(num)
proposals = proposals[topk_inds, :]
return proposals
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment