Focal Loss for Dense Object Detection：RetinaNet

RetinaNet

mmdetection配置文件分析

1.retinanet_r50_fpn_1x_coco.py

model = dict(
    type='RetinaNet',  # 检测器名称
    pretrained='torchvision://resnet50',  # 基于ImageNet的预训练模型
    backbone=dict(  # 骨架配置信息
        type='ResNet',  # 骨架网络名称
        depth=50,  # 骨架网络的深度
        num_stages=4,  # 骨架网络的阶段数（一般提取特征层从阶段集合中提取）
        out_indices=(0, 1, 2, 3),  # 提取特征图层下标
        frozen_stages=1,  # 骨架第一阶段权重系数被冻结
        norm_cfg=dict(  # 归一化层的配置
            type='BN',  # 归一化层的类型,一般有BN和GN两种
            requires_grad=True),  # 是否训练归一化层权重
        norm_eval=True,  # Whether to freeze the statistics in BN
        style='pytorch'),
    # The style of backbone, 'pytorch' means that stride 2 layers are in 3x3 conv, 'caffe' means stride 2 layers are in 1x1 convs.
    neck=dict(  # 检测器网络颈部
        type='FPN',  # FPN特征金字塔还有NASFPN,PAFPN等等
        in_channels=[256, 512, 1024, 2048],  # FPN特征融合金字塔的输入通道与backbone.out_indices应该一致
        out_channels=256,  # 特征金字塔每层输出的通道数
        start_level=1,
        add_extra_convs='on_input',
        num_outs=5),  # The number of output scales
    bbox_head=dict(  # Config of box head in the RetinaHead.
        type='RetinaHead',
        num_classes=80,  # 类别数
        in_channels=256,  # 输入通道
        stacked_convs=4, 
        feat_channels=256,
        anchor_generator=dict(
            type='AnchorGenerator',
            octave_base_scale=4,
            scales_per_octave=3,
            ratios=[0.5, 1.0, 2.0],
            strides=[8, 16, 32, 64, 128]),
        bbox_coder=dict(
            type='DeltaXYWHBBoxCoder',
            target_means=[0.0, 0.0, 0.0, 0.0],
            target_stds=[1.0, 1.0, 1.0, 1.0]),
        loss_cls=dict(
            type='FocalLoss',
            use_sigmoid=True,
            gamma=2.0,
            alpha=0.25,
            loss_weight=1.0),
        loss_bbox=dict(type='L1Loss', loss_weight=1.0)))
train_cfg = dict(
    assigner=dict(
        type='MaxIoUAssigner',
        pos_iou_thr=0.5,
        neg_iou_thr=0.4,
        min_pos_iou=0,
        ignore_iof_thr=-1),
    allowed_border=-1,
    pos_weight=-1,
    debug=False)
test_cfg = dict(
    nms_pre=1000,
    min_bbox_size=0,
    score_thr=0.05,
    nms=dict(type='nms', iou_threshold=0.5),
    max_per_img=100)
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='LoadAnnotations', with_bbox=True),
    dict(type='Resize', img_scale=(1333, 800), keep_ratio=True),
    dict(type='RandomFlip', flip_ratio=0.5),
    dict(
        type='Normalize',
        mean=[123.675, 116.28, 103.53],
        std=[58.395, 57.12, 57.375],
        to_rgb=True),
    dict(type='Pad', size_divisor=32),
    dict(type='DefaultFormatBundle'),
    dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels'])
]
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',
                mean=[123.675, 116.28, 103.53],
                std=[58.395, 57.12, 57.375],
                to_rgb=True),
            dict(type='Pad', size_divisor=32),
            dict(type='ImageToTensor', keys=['img']),
            dict(type='Collect', keys=['img'])
        ])
]
data = dict(
    samples_per_gpu=2,
    workers_per_gpu=1,
    train=dict(
        type='CocoDataset',
        ann_file='../data/coco/annotations/instances_val2017.json',
        img_prefix='../data/coco/val2017/',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(type='LoadAnnotations', with_bbox=True),
            dict(type='Resize', img_scale=(1333, 800), keep_ratio=True),
            dict(type='RandomFlip', flip_ratio=0.5),
            dict(
                type='Normalize',
                mean=[123.675, 116.28, 103.53],
                std=[58.395, 57.12, 57.375],
                to_rgb=True),
            dict(type='Pad', size_divisor=32),
            dict(type='DefaultFormatBundle'),
            dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels'])
        ]),
    val=dict(
        type='CocoDataset',
        ann_file='../data/coco/annotations/instances_val2017.json',
        img_prefix='../data/coco/val2017/',
        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',
                        mean=[123.675, 116.28, 103.53],
                        std=[58.395, 57.12,