Skip to content

GitLab

Commit e907139c authored by GothicAi's avatar GothicAi Committed by Kai Chen
Browse files

add instaboost in coco.py (#1739) 



* add instaboost in coco.py

* Revert "add instaboost in coco.py"

This reverts commit 51af81ef88991782d54750761ef2dd394c7a38a8.

* add instaboost in data pipeline

* code format

* Update requirements.txt

* add README.md

* bug fix

* add link

* Update README.md

* install instaboost when necessary, keep r50 cfgs only

* ready to merge

* fix typo

* pass yapf

* minor formats
Co-authored-by: default avatarWenwei Zhang <40779233+ZwwWayne@users.noreply.github.com>
Co-authored-by: default avatarKai Chen <chenkaidev@gmail.com>
parent da6b1c82
Hide whitespace changes
Inline Side-by-side
Showing with 758 additions and 1 deletion
README.md
View file @ e907139c
......@@ -80,6 +80,7 @@ Other features
- [x] Generalized Attention
- [x] GCNet
- [x] Mixed Precision (FP16) Training
- [x] [InstaBoost](configs/instaboost/README.md)
## Installation
......
configs/instaboost/README.md 0 → 100644
View file @ e907139c
# InstaBoost for MMDetection
Configs in this directory is the implementation for ICCV2019 paper "InstaBoost: Boosting Instance Segmentation Via Probability Map Guided Copy-Pasting" and provided by the authors of the paper. InstaBoost is a data augmentation method for object detection and instance segmentation. The paper has been released on [`arXiv`](https://arxiv.org/abs/1908.07801).
```
@inproceedings{fang2019instaboost,
title={Instaboost: Boosting instance segmentation via probability map guided copy-pasting},
author={Fang, Hao-Shu and Sun, Jianhua and Wang, Runzhong and Gou, Minghao and Li, Yong-Lu and Lu, Cewu},
booktitle={Proceedings of the IEEE International Conference on Computer Vision},
pages={682--691},
year={2019}
}
```
## Usage
### Requirements
You need to install `instaboostfast` before using it.
```
pip install instaboostfast
```
The code and more details can be found [here](https://github.com/GothicAi/Instaboost).
### Integration with MMDetection
InstaBoost have been already integrated in the data pipeline, thus all you need is to add or change **InstaBoost** configurations after **LoadImageFromFile**. We have provided examples like [this](mask_rcnn_r50_fpn_instaboost_4x.py#L121). You can refer to [`InstaBoostConfig`](https://github.com/GothicAi/InstaBoost-pypi#instaboostconfig) for more details.
## Results and Models
- All models were trained on `coco_2017_train` and tested on `coco_2017_val` for conveinience of evaluation and comparison. In the paper, the results are obtained from `test-dev`.
- To balance accuracy and training time when using InstaBoost, models released in this page are all trained for 48 Epochs. Other training and testing configs strictly follow the original framework.
- The results and models are provided by the [authors](https://github.com/GothicAi/Instaboost) (many thanks).
| InstaBoost | Network | Backbone | Lr schd | box AP | mask AP | Download |
| :-------------: | :-------------: | :--------: | :-----: | :----: | :-----: | :-----------------: |
| × | Mask R-CNN | R-50-FPN | 1x | 37.3 | 34.2 | - |
| √ | Mask R-CNN | R-50-FPN | 4x |**40.0**|**36.2** |[Baidu](https://pan.baidu.com/s/1PLn1K5qreDoM4wh7nbsLqA) / [Google](https://drive.google.com/file/d/1uUT1qc3oYS8xHLyM7bJWgxBNbW-9sa1f/view?usp=sharing)|
| × | Mask R-CNN | R-101-FPN | 1x | 39.4 | 35.9 | - |
| √ | Mask R-CNN | R-101-FPN | 4x |**42.1**|**37.8** |[Baidu](https://pan.baidu.com/s/1IZpqCDrcrOiwNJ-Y_3wpOQ) / [Google](https://drive.google.com/file/d/1idGMPexovIDUHXSNlpIA1mjKzgnFrcW3/view?usp=sharing)|
| × | Mask R-CNN | X-101-64x4d-FPN | 1x | 42.1 | 38.0 | - |
| × | Mask R-CNN | X-101-64x4d-FPN | 2x | *42.0* | *37.7* | - |
| √ | Mask R-CNN | X-101-64x4d-FPN | 4x |**44.5**|**39.5** |[Baidu](https://pan.baidu.com/s/1KrHQBHcHjWONpXbC2qUzxw) / [Google](https://drive.google.com/file/d/1qD4V9uYbtpaZBmTMTgP7f0uw46zroY9-/view?usp=sharing)|
| × | Cascade R-CNN | R-101-FPN | 1x | 42.6 | 37.0 | - |
| √ | Cascade R-CNN | R-101-FPN | 4x |**45.4**|**39.2** |[Baidu](https://pan.baidu.com/s/1_4cJ0B9fugcA-oBHYe9o_A) / [Google](https://drive.google.com/file/d/1xhiuFoOMQyDIvOrz6MiAZPboRRe1YK8p/view?usp=sharing)|
| × | Cascade R-CNN | X-101-64x4d-FPN | 1x | 45.4 | 39.1 | - |
| √ | Cascade R-CNN | X-101-64x4d-FPN | 4x |**47.2**|**40.4** |[Baidu](https://pan.baidu.com/s/1nu73IpRbTEb4caPMHWJMXA) / [Google](https://drive.google.com/file/d/11iaKH-ZeVCi-65wzlT5OxxUOkREMzXRW/view?usp=sharing)|
| × | SSD | VGG16-512 | 120e | 29.3 | - | - |
| √ | SSD | VGG16-512 | 360e |**30.3**| - |[Baidu](https://pan.baidu.com/s/1G-1atZ81A8mLLx8taJAuwQ) / [Google](https://drive.google.com/file/d/1sqMIEusZw2Y7Ge8DuJgmhSP-2V74BNKy/view?usp=sharing)|
configs/instaboost/cascade_mask_rcnn_r50_fpn_instaboost_4x.py 0 → 100644
View file @ e907139c
# model settings
model = dict(
type='CascadeRCNN',
num_stages=3,
pretrained='torchvision://resnet50',
backbone=dict(
type='ResNet',
depth=50,
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='RPNHead',
in_channels=256,
feat_channels=256,
anchor_scales=[8],
anchor_ratios=[0.5, 1.0, 2.0],
anchor_strides=[4, 8, 16, 32, 64],
target_means=[.0, .0, .0, .0],
target_stds=[1.0, 1.0, 1.0, 1.0],
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0 / 9.0, 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=[0.1, 0.1, 0.2, 0.2],
reg_class_agnostic=True,
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)),
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=True,
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)),
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=[0.033, 0.033, 0.067, 0.067],
reg_class_agnostic=True,
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0))
],
mask_roi_extractor=dict(
type='SingleRoIExtractor',
roi_layer=dict(type='RoIAlign', out_size=14, sample_num=2),
out_channels=256,
featmap_strides=[4, 8, 16, 32]),
mask_head=dict(
type='FCNMaskHead',
num_convs=4,
in_channels=256,
conv_out_channels=256,
num_classes=81,
loss_mask=dict(
type='CrossEntropyLoss', use_mask=True, loss_weight=1.0)))
# model training and testing settings
train_cfg = dict(
rpn=dict(
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=0,
pos_weight=-1,
debug=False),
rpn_proposal=dict(
nms_across_levels=False,
nms_pre=2000,
nms_post=2000,
max_num=2000,
nms_thr=0.7,
min_bbox_size=0),
rcnn=[
dict(
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.5,
neg_iou_thr=0.5,
min_pos_iou=0.5,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=512,
pos_fraction=0.25,
neg_pos_ub=-1,
add_gt_as_proposals=True),
mask_size=28,
pos_weight=-1,
debug=False),
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=512,
pos_fraction=0.25,
neg_pos_ub=-1,
add_gt_as_proposals=True),
mask_size=28,
pos_weight=-1,
debug=False),
dict(
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.7,
neg_iou_thr=0.7,
min_pos_iou=0.7,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=512,
pos_fraction=0.25,
neg_pos_ub=-1,
add_gt_as_proposals=True),
mask_size=28,
pos_weight=-1,
debug=False)
],
stage_loss_weights=[1, 0.5, 0.25])
test_cfg = dict(
rpn=dict(
nms_across_levels=False,
nms_pre=1000,
nms_post=1000,
max_num=1000,
nms_thr=0.7,
min_bbox_size=0),
rcnn=dict(
score_thr=0.05,
nms=dict(type='nms', iou_thr=0.5),
max_per_img=100,
mask_thr_binary=0.5))
# 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)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='InstaBoost',
action_candidate=('normal', 'horizontal', 'skip'),
action_prob=(1, 0, 0),
scale=(0.8, 1.2),
dx=15,
dy=15,
theta=(-1, 1),
color_prob=0.5,
hflag=False,
aug_ratio=0.5),
dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
dict(type='Resize', img_scale=(1333, 800), keep_ratio=True),
dict(type='RandomFlip', flip_ratio=0.5),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='DefaultFormatBundle'),
dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks']),
]
test_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='MultiScaleFlipAug',
img_scale=(1333, 800),
flip=False,
transforms=[
dict(type='Resize', keep_ratio=True),
dict(type='RandomFlip'),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='ImageToTensor', keys=['img']),
dict(type='Collect', keys=['img']),
])
]
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/',
pipeline=train_pipeline),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
pipeline=test_pipeline),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
pipeline=test_pipeline))
# 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=[32, 44])
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 = 48
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/cascade_mask_rcnn_r50_fpn_instaboost_4x'
load_from = None
resume_from = None
workflow = [('train', 1)]
configs/instaboost/mask_rcnn_r50_fpn_instaboost_4x.py 0 → 100644
View file @ e907139c
# model settings
model = dict(
type='MaskRCNN',
pretrained='torchvision://resnet50',
backbone=dict(
type='ResNet',
depth=50,
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='RPNHead',
in_channels=256,
feat_channels=256,
anchor_scales=[8],
anchor_ratios=[0.5, 1.0, 2.0],
anchor_strides=[4, 8, 16, 32, 64],
target_means=[.0, .0, .0, .0],
target_stds=[1.0, 1.0, 1.0, 1.0],
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0 / 9.0, 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=[0.1, 0.1, 0.2, 0.2],
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)),
mask_roi_extractor=dict(
type='SingleRoIExtractor',
roi_layer=dict(type='RoIAlign', out_size=14, sample_num=2),
out_channels=256,
featmap_strides=[4, 8, 16, 32]),
mask_head=dict(
type='FCNMaskHead',
num_convs=4,
in_channels=256,
conv_out_channels=256,
num_classes=81,
loss_mask=dict(
type='CrossEntropyLoss', use_mask=True, loss_weight=1.0)))
# model training and testing settings
train_cfg = dict(
rpn=dict(
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=0,
pos_weight=-1,
debug=False),
rpn_proposal=dict(
nms_across_levels=False,
nms_pre=2000,
nms_post=2000,
max_num=2000,
nms_thr=0.7,
min_bbox_size=0),
rcnn=dict(
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.5,
neg_iou_thr=0.5,
min_pos_iou=0.5,
ignore_iof_thr=-1),
sampler=dict(
type='RandomSampler',
num=512,
pos_fraction=0.25,
neg_pos_ub=-1,
add_gt_as_proposals=True),
mask_size=28,
pos_weight=-1,
debug=False))
test_cfg = dict(
rpn=dict(
nms_across_levels=False,
nms_pre=1000,
nms_post=1000,
max_num=1000,
nms_thr=0.7,
min_bbox_size=0),
rcnn=dict(
score_thr=0.05,
nms=dict(type='nms', iou_thr=0.5),
max_per_img=100,
mask_thr_binary=0.5))
# 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)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='InstaBoost',
action_candidate=('normal', 'horizontal', 'skip'),
action_prob=(1, 0, 0),
scale=(0.8, 1.2),
dx=15,
dy=15,
theta=(-1, 1),
color_prob=0.5,
hflag=False,
aug_ratio=0.5),
dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
dict(type='Resize', img_scale=(1333, 800), keep_ratio=True),
dict(type='RandomFlip', flip_ratio=0.5),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='DefaultFormatBundle'),
dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks']),
]
test_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='MultiScaleFlipAug',
img_scale=(1333, 800),
flip=False,
transforms=[
dict(type='Resize', keep_ratio=True),
dict(type='RandomFlip'),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='ImageToTensor', keys=['img']),
dict(type='Collect', keys=['img']),
])
]
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/',
pipeline=train_pipeline),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
pipeline=test_pipeline),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
pipeline=test_pipeline))
# 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=[32, 44])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
interval=50,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook')
])
# yapf:enable
evaluation = dict(interval=1)
# runtime settings
total_epochs = 48
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/mask_rcnn_r50_fpn_instaboost_4x'
load_from = None
resume_from = None
workflow = [('train', 1)]
configs/instaboost/ssd300_coco_instaboost_4x.py 0 → 100644
View file @ e907139c
# model settings
input_size = 300
model = dict(
type='SingleStageDetector',
pretrained='open-mmlab://vgg16_caffe',
backbone=dict(
type='SSDVGG',
input_size=input_size,
depth=16,
with_last_pool=False,
ceil_mode=True,
out_indices=(3, 4),
out_feature_indices=(22, 34),
l2_norm_scale=20),
neck=None,
bbox_head=dict(
type='SSDHead',
input_size=input_size,
in_channels=(512, 1024, 512, 256, 256, 256),
num_classes=81,
anchor_strides=(8, 16, 32, 64, 100, 300),
basesize_ratio_range=(0.15, 0.9),
anchor_ratios=([2], [2, 3], [2, 3], [2, 3], [2], [2]),
target_means=(.0, .0, .0, .0),
target_stds=(0.1, 0.1, 0.2, 0.2)))
cudnn_benchmark = True
train_cfg = dict(
assigner=dict(
type='MaxIoUAssigner',
pos_iou_thr=0.5,
neg_iou_thr=0.5,
min_pos_iou=0.,
ignore_iof_thr=-1,
gt_max_assign_all=False),
smoothl1_beta=1.,
allowed_border=-1,
pos_weight=-1,
neg_pos_ratio=3,
debug=False)
test_cfg = dict(
nms=dict(type='nms', iou_thr=0.45),
min_bbox_size=0,
score_thr=0.02,
max_per_img=200)
# model training and testing settings
# dataset settings
dataset_type = 'CocoDataset'
data_root = 'data/coco/'
img_norm_cfg = dict(mean=[123.675, 116.28, 103.53], std=[1, 1, 1], to_rgb=True)
train_pipeline = [
dict(type='LoadImageFromFile', to_float32=True),
dict(
type='InstaBoost',
action_candidate=('normal', 'horizontal', 'skip'),
action_prob=(1, 0, 0),
scale=(0.8, 1.2),
dx=15,
dy=15,
theta=(-1, 1),
color_prob=0.5,
hflag=False,
aug_ratio=0.5),
dict(type='LoadAnnotations', with_bbox=True),
dict(
type='PhotoMetricDistortion',
brightness_delta=32,
contrast_range=(0.5, 1.5),
saturation_range=(0.5, 1.5),
hue_delta=18),
dict(
type='Expand',
mean=img_norm_cfg['mean'],
to_rgb=img_norm_cfg['to_rgb'],
ratio_range=(1, 4)),
dict(
type='MinIoURandomCrop',
min_ious=(0.1, 0.3, 0.5, 0.7, 0.9),
min_crop_size=0.3),
dict(type='Resize', img_scale=(300, 300), keep_ratio=False),
dict(type='Normalize', **img_norm_cfg),
dict(type='RandomFlip', flip_ratio=0.5),
dict(type='DefaultFormatBundle'),
dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']),
]
test_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='MultiScaleFlipAug',
img_scale=(300, 300),
flip=False,
transforms=[
dict(type='Resize', keep_ratio=False),
dict(type='Normalize', **img_norm_cfg),
dict(type='ImageToTensor', keys=['img']),
dict(type='Collect', keys=['img']),
])
]
data = dict(
imgs_per_gpu=8,
workers_per_gpu=3,
train=dict(
type='RepeatDataset',
times=5,
dataset=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_train2017.json',
img_prefix=data_root + 'train2017/',
pipeline=train_pipeline)),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
pipeline=test_pipeline),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
pipeline=test_pipeline))
# optimizer
optimizer = dict(type='SGD', lr=2e-3, momentum=0.9, weight_decay=5e-4)
optimizer_config = dict()
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[64, 88])
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 = 96
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/ssd300_coco_instaboost_4x'
load_from = None
resume_from = None
workflow = [('train', 1)]
docs/MODEL_ZOO.md
View file @ e907139c
......@@ -230,6 +230,10 @@ Please refer to [Weight Standardization](https://github.com/open-mmlab/mmdetecti
Please refer to [Deformable Convolutional Networks](https://github.com/open-mmlab/mmdetection/blob/master/configs/dcn) for details.
### Instaboost
Please refer to [Instaboost](https://github.com/open-mmlab/mmdetection/blob/master/configs/instaboost) for details.
### Libra R-CNN
Please refer to [Libra R-CNN](https://github.com/open-mmlab/mmdetection/blob/master/configs/libra_rcnn) for details.
......
mmdet/datasets/pipelines/__init__.py
View file @ e907139c
from .compose import Compose
from .formating import (Collect, ImageToTensor, ToDataContainer, ToTensor,
Transpose, to_tensor)
from .instaboost import InstaBoost
from .loading import LoadAnnotations, LoadImageFromFile, LoadProposals
from .test_aug import MultiScaleFlipAug
from .transforms import (Albu, Expand, MinIoURandomCrop, Normalize, Pad,
......@@ -12,5 +13,5 @@ __all__ = [
'Transpose', 'Collect', 'LoadAnnotations', 'LoadImageFromFile',
'LoadProposals', 'MultiScaleFlipAug', 'Resize', 'RandomFlip', 'Pad',
'RandomCrop', 'Normalize', 'SegRescale', 'MinIoURandomCrop', 'Expand',
'PhotoMetricDistortion', 'Albu'
'PhotoMetricDistortion', 'Albu', 'InstaBoost'
]
mmdet/datasets/pipelines/instaboost.py 0 → 100644
View file @ e907139c
import numpy as np
from ..registry import PIPELINES
@PIPELINES.register_module
class InstaBoost(object):
"""
Data augmentation method in paper "InstaBoost: Boosting Instance
Segmentation Via Probability Map Guided Copy-Pasting"
Implementation details can refer to https://github.com/GothicAi/Instaboost.
"""
def __init__(self,
action_candidate=('normal', 'horizontal', 'skip'),
action_prob=(1, 0, 0),
scale=(0.8, 1.2),
dx=15,
dy=15,
theta=(-1, 1),
color_prob=0.5,
hflag=False,
aug_ratio=0.5):
try:
import instaboostfast as instaboost
except ImportError:
raise ImportError(
'Please run "pip install instaboostfast" '
'to install instaboostfast first for instaboost augmentation.')
self.cfg = instaboost.InstaBoostConfig(action_candidate, action_prob,
scale, dx, dy, theta,
color_prob, hflag)
self.aug_ratio = aug_ratio
def _load_anns(self, results):
labels = results['ann_info']['labels']
masks = results['ann_info']['masks']
bboxes = results['ann_info']['bboxes']
n = len(labels)
anns = []
for i in range(n):
label = labels[i]
bbox = bboxes[i]
mask = masks[i]
x1, y1, x2, y2 = bbox
bbox = [x1, y1, x2 - x1 + 1, y2 - y1 + 1]
anns.append({
'category_id': label,
'segmentation': mask,
'bbox': bbox
})
return anns
def _parse_anns(self, results, anns, img):
gt_bboxes = []
gt_labels = []
gt_masks_ann = []
for ann in anns:
x1, y1, w, h = ann['bbox']
bbox = [x1, y1, x1 + w - 1, y1 + h - 1]
gt_bboxes.append(bbox)
gt_labels.append(ann['category_id'])
gt_masks_ann.append(ann['segmentation'])
gt_bboxes = np.array(gt_bboxes, dtype=np.float32)
gt_labels = np.array(gt_labels, dtype=np.int64)
results['ann_info']['labels'] = gt_labels
results['ann_info']['bboxes'] = gt_bboxes
results['ann_info']['masks'] = gt_masks_ann
results['img'] = img
return results
def __call__(self, results):
img = results['img']
anns = self._load_anns(results)
if np.random.choice([0, 1], p=[1 - self.aug_ratio, self.aug_ratio]):
try:
import instaboostfast as instaboost
except ImportError:
raise ImportError('Please run "pip install instaboostfast" '
'to install instaboostfast first.')
anns, img = instaboost.get_new_data(
anns, img, self.cfg, background=None)
results = self._parse_anns(results, anns, img)
return results
def __repr__(self):
repr_str = self.__class__.__name__
repr_str += ('(cfg={}, aug_ratio={})').format(self.cfg, self.aug_ratio)
return repr_str
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