【YOLO改进】换遍IoU损失函数之EIoU Loss（基于MMYOLO）

EIoU损失函数

设计原理

一、IoU的局限性

IoU（Intersection over Union）是一种常用于评估目标检测模型性能的指标，特别是在计算预测边界框与真实边界框之间的重叠程度时。然而，IoU存在一些局限性，尤其是当两个边界框没有任何交集时，IoU 的值为0，这使得梯度更新停滞，不利于模型的进一步学习和优化。

二、EIoU的引入

为了解决这一问题，引入了EIoU（Enhanced Intersection over Union）损失函数。EIoU 不仅考虑了边界框间的重叠区域，还引入了其他度量，如边界框中心点的距离，以及边界框的宽度和高度的相对差异。这样的设计使得即使两个边界框不重叠，损失函数仍然可以提供有效的梯度，从而促进模型的训练和收敛。

计算步骤

一、计算IoU

计算两个边界框A和B的交集面积I。
计算两个边界框的并集面积U。
IoU计算公式为：I/U

二、计算中心点距离的公式

中心点距离是预测框和真实框中心点之间的欧氏距离。设预测框中心为 $(x_p,y_p)$ ，真实框中心为 $(x_g,y_g)$ ，则中心距离 $D_c$ 计算为：

$D_c = \sqrt{(x_p - x_g)^2 + (y_p - y_g)^2}$

三、计算宽高比的差异

宽度差异 $w_{diff}$ 和高度差异 $h_{diff}$ 分别为预测框和真实框宽度和高度的相对差值。计算方法可以是简单的差值或者比例差等。

四、整合以上度量

EIoU将上述度量整合到一个损失函数中，通常形式为：

$\text{EIoU Loss} = 1 - \text{IoU} + \lambda_1 D_c + \lambda_2 (w_{\text{diff}} + h_{\text{diff}})$

其中， $\lambda_1$ 和 $\lambda_2$ 是调节中心距离和宽高差异影响的超参数。

使用PyTorch实现EIoU计算的源代码

import torch
import torch.nn.functional as Fdef bbox_iou(boxes1, boxes2):"""计算两组边界框的IoU。boxes1, boxes2: [N, 4] (x1, y1, x2, y2)"""area1 = (boxes1[:, 2] - boxes1[:, 0]) * (boxes1[:, 3] - boxes1[:, 1])area2 = (boxes2[:, 2] - boxes2[:, 0]) * (boxes2[:, 3] - boxes2[:, 1])inter_x1 = torch.max(boxes1[:, 0], boxes2[:, 0])inter_y1 = torch.max(boxes1[:, 1], boxes2[:, 1])inter_x2 = torch.min(boxes1[:, 2], boxes2[:, 2])inter_y2 = torch.min(boxes1[:, 3], boxes2[:, 3])inter_area = torch.clamp(inter_x2 - inter_x1, min=0) * torch.clamp(inter_y2 - inter_y1, min=0)union_area = area1 + area2 - inter_areareturn inter_area / union_areadef eiou_loss(pred_boxes, target_boxes, lambda1=1, lambda2=1):"""计算EIoU损失。pred_boxes, target_boxes: [N, 4] (x1, y1, x2, y2)"""iou = bbox_iou(pred_boxes, target_boxes)# 计算中心点center_pred = (pred_boxes[:, :2] + pred_boxes[:, 2:4]) / 2center_target = (target_boxes[:, :2] + target_boxes[:, 2:4]) / 2# 计算中心点距离dc = torch.sqrt(torch.sum((center_pred - center_target) ** 2, dim=1))# 计算宽高差异wh_pred = pred_boxes[:, 2:4] - pred_boxes[:, :2]wh_target = target_boxes[:, 2:4] - target_boxes[:, :2]wh_diff = torch.abs(wh_pred - wh_target).sum(dim=1)# 计算EIoU损失loss = 1 - iou + lambda1 * dc + lambda2 * wh_diffreturn loss.mean()# 示例用法
pred_boxes = torch.tensor([[25, 25, 75, 75], [50, 50, 100, 100]], dtype=torch.float32)
target_boxes = torch.tensor([[30, 30, 70, 70], [70, 70, 120, 120]], dtype=torch.float32)loss = eiou_loss(pred_boxes, target_boxes)
print(f"EIoU Loss: {loss}")

替换EIoU损失函数(基于MMYOLO)

由于MMYOLO中没有实现EIoU损失函数，所以需要在mmyolo/models/iou_loss.py中添加EIoU的计算和对应的iou_mode，修改完以后在终端运行

python setup.py install

再在配置文件中进行修改即可。修改例子如下：

    elif iou_mode == "eiou":# CIoU = IoU - ( (ρ^2(b_pred,b_gt) / c^2) + (alpha x v) )# calculate enclose area (c^2)enclose_area = enclose_w**2 + enclose_h**2 + eps# calculate ρ^2(b_pred,b_gt):# euclidean distance between b_pred(bbox2) and b_gt(bbox1)# center point, because bbox format is xyxy -> left-top xy and# right-bottom xy, so need to / 4 to get center point.rho2_left_item = ((bbox2_x1 + bbox2_x2) - (bbox1_x1 + bbox1_x2))**2 / 4rho2_right_item = ((bbox2_y1 + bbox2_y2) -(bbox1_y1 + bbox1_y2))**2 / 4rho2 = rho2_left_item + rho2_right_item  # rho^2 (ρ^2)rho_w2 = ((bbox2_x2 - bbox2_x1) - (bbox1_x2 - bbox1_x1)) ** 2rho_h2 = ((bbox2_y2 - bbox2_y1) - (bbox1_y2 - bbox1_y1)) ** 2cw2 = enclose_w ** 2 + epsch2 = enclose_h ** 2 + epsious = ious - (rho2 / enclose_area + rho_w2 / cw2 + rho_h2 / ch2)

修改后的配置文件（以configs/yolov5/yolov5_s-v61_syncbn_8xb16-300e_coco.py为例）

_base_ = ['../_base_/default_runtime.py', '../_base_/det_p5_tta.py']# ========================Frequently modified parameters======================
# -----data related-----
data_root = 'data/coco/'  # Root path of data
# Path of train annotation file
train_ann_file = 'annotations/instances_train2017.json'
train_data_prefix = 'train2017/'  # Prefix of train image path
# Path of val annotation file
val_ann_file = 'annotations/instances_val2017.json'
val_data_prefix = 'val2017/'  # Prefix of val image pathnum_classes = 80  # Number of classes for classification
# Batch size of a single GPU during training
train_batch_size_per_gpu = 16
# Worker to pre-fetch data for each single GPU during training
train_num_workers = 8
# persistent_workers must be False if num_workers is 0
persistent_workers = True# -----model related-----
# Basic size of multi-scale prior box
anchors = [[(10, 13), (16, 30), (33, 23)],  # P3/8[(30, 61), (62, 45), (59, 119)],  # P4/16[(116, 90), (156, 198), (373, 326)]  # P5/32
]# -----train val related-----
# Base learning rate for optim_wrapper. Corresponding to 8xb16=128 bs
base_lr = 0.01
max_epochs = 300  # Maximum training epochsmodel_test_cfg = dict(# The config of multi-label for multi-class prediction.multi_label=True,# The number of boxes before NMSnms_pre=30000,score_thr=0.001,  # Threshold to filter out boxes.nms=dict(type='nms', iou_threshold=0.65),  # NMS type and thresholdmax_per_img=300)  # Max number of detections of each image# ========================Possible modified parameters========================
# -----data related-----
img_scale = (640, 640)  # width, height
# Dataset type, this will be used to define the dataset
dataset_type = 'YOLOv5CocoDataset'
# Batch size of a single GPU during validation
val_batch_size_per_gpu = 1
# Worker to pre-fetch data for each single GPU during validation
val_num_workers = 2# Config of batch shapes. Only on val.
# It means not used if batch_shapes_cfg is None.
batch_shapes_cfg = dict(type='BatchShapePolicy',batch_size=val_batch_size_per_gpu,img_size=img_scale[0],# The image scale of padding should be divided by pad_size_divisorsize_divisor=32,# Additional paddings for pixel scaleextra_pad_ratio=0.5)# -----model related-----
# The scaling factor that controls the depth of the network structure
deepen_factor = 0.33
# The scaling factor that controls the width of the network structure
widen_factor = 0.5
# Strides of multi-scale prior box
strides = [8, 16, 32]
num_det_layers = 3  # The number of model output scales
norm_cfg = dict(type='BN', momentum=0.03, eps=0.001)  # Normalization config# -----train val related-----
affine_scale = 0.5  # YOLOv5RandomAffine scaling ratio
loss_cls_weight = 0.5
loss_bbox_weight = 0.05
loss_obj_weight = 1.0
prior_match_thr = 4.  # Priori box matching threshold
# The obj loss weights of the three output layers
obj_level_weights = [4., 1., 0.4]
lr_factor = 0.01  # Learning rate scaling factor
weight_decay = 0.0005
# Save model checkpoint and validation intervals
save_checkpoint_intervals = 10
# The maximum checkpoints to keep.
max_keep_ckpts = 3
# Single-scale training is recommended to
# be turned on, which can speed up training.
env_cfg = dict(cudnn_benchmark=True)# ===============================Unmodified in most cases====================
model = dict(type='YOLODetector',data_preprocessor=dict(type='mmdet.DetDataPreprocessor',mean=[0., 0., 0.],std=[255., 255., 255.],bgr_to_rgb=True),backbone=dict(##使用YOLOv8的主干网络type='YOLOv8CSPDarknet',deepen_factor=deepen_factor,widen_factor=widen_factor,norm_cfg=norm_cfg,act_cfg=dict(type='SiLU', inplace=True)),neck=dict(type='YOLOv5PAFPN',deepen_factor=deepen_factor,widen_factor=widen_factor,in_channels=[256, 512, 1024],out_channels=[256, 512, 1024],num_csp_blocks=3,norm_cfg=norm_cfg,act_cfg=dict(type='SiLU', inplace=True)),bbox_head=dict(type='YOLOv5Head',head_module=dict(type='YOLOv5HeadModule',num_classes=num_classes,in_channels=[256, 512, 1024],widen_factor=widen_factor,featmap_strides=strides,num_base_priors=3),prior_generator=dict(type='mmdet.YOLOAnchorGenerator',base_sizes=anchors,strides=strides),# scaled based on number of detection layersloss_cls=dict(type='mmdet.CrossEntropyLoss',use_sigmoid=True,reduction='mean',loss_weight=loss_cls_weight *(num_classes / 80 * 3 / num_det_layers)),# 修改此处实现IoU损失函数的替换loss_bbox=dict(type='IoULoss',iou_mode='eiou',bbox_format='xywh',eps=1e-7,reduction='mean',loss_weight=loss_bbox_weight * (3 / num_det_layers),return_iou=True),loss_obj=dict(type='mmdet.CrossEntropyLoss',use_sigmoid=True,reduction='mean',loss_weight=loss_obj_weight *((img_scale[0] / 640)**2 * 3 / num_det_layers)),prior_match_thr=prior_match_thr,obj_level_weights=obj_level_weights),test_cfg=model_test_cfg)albu_train_transforms = [dict(type='Blur', p=0.01),dict(type='MedianBlur', p=0.01),dict(type='ToGray', p=0.01),dict(type='CLAHE', p=0.01)
]pre_transform = [dict(type='LoadImageFromFile', file_client_args=_base_.file_client_args),dict(type='LoadAnnotations', with_bbox=True)
]train_pipeline = [*pre_transform,dict(type='Mosaic',img_scale=img_scale,pad_val=114.0,pre_transform=pre_transform),dict(type='YOLOv5RandomAffine',max_rotate_degree=0.0,max_shear_degree=0.0,scaling_ratio_range=(1 - affine_scale, 1 + affine_scale),# img_scale is (width, height)border=(-img_scale[0] // 2, -img_scale[1] // 2),border_val=(114, 114, 114)),dict(type='mmdet.Albu',transforms=albu_train_transforms,bbox_params=dict(type='BboxParams',format='pascal_voc',label_fields=['gt_bboxes_labels', 'gt_ignore_flags']),keymap={'img': 'image','gt_bboxes': 'bboxes'}),dict(type='YOLOv5HSVRandomAug'),dict(type='mmdet.RandomFlip', prob=0.5),dict(type='mmdet.PackDetInputs',meta_keys=('img_id', 'img_path', 'ori_shape', 'img_shape', 'flip','flip_direction'))
]train_dataloader = dict(batch_size=train_batch_size_per_gpu,num_workers=train_num_workers,persistent_workers=persistent_workers,pin_memory=True,sampler=dict(type='DefaultSampler', shuffle=True),dataset=dict(type=dataset_type,data_root=data_root,ann_file=train_ann_file,data_prefix=dict(img=train_data_prefix),filter_cfg=dict(filter_empty_gt=False, min_size=32),pipeline=train_pipeline))test_pipeline = [dict(type='LoadImageFromFile', file_client_args=_base_.file_client_args),dict(type='YOLOv5KeepRatioResize', scale=img_scale),dict(type='LetterResize',scale=img_scale,allow_scale_up=False,pad_val=dict(img=114)),dict(type='LoadAnnotations', with_bbox=True, _scope_='mmdet'),dict(type='mmdet.PackDetInputs',meta_keys=('img_id', 'img_path', 'ori_shape', 'img_shape','scale_factor', 'pad_param'))
]val_dataloader = dict(batch_size=val_batch_size_per_gpu,num_workers=val_num_workers,persistent_workers=persistent_workers,pin_memory=True,drop_last=False,sampler=dict(type='DefaultSampler', shuffle=False),dataset=dict(type=dataset_type,data_root=data_root,test_mode=True,data_prefix=dict(img=val_data_prefix),ann_file=val_ann_file,pipeline=test_pipeline,batch_shapes_cfg=batch_shapes_cfg))test_dataloader = val_dataloaderparam_scheduler = None
optim_wrapper = dict(type='OptimWrapper',optimizer=dict(type='SGD',lr=base_lr,momentum=0.937,weight_decay=weight_decay,nesterov=True,batch_size_per_gpu=train_batch_size_per_gpu),constructor='YOLOv5OptimizerConstructor')default_hooks = dict(param_scheduler=dict(type='YOLOv5ParamSchedulerHook',scheduler_type='linear',lr_factor=lr_factor,max_epochs=max_epochs),checkpoint=dict(type='CheckpointHook',interval=save_checkpoint_intervals,save_best='auto',max_keep_ckpts=max_keep_ckpts))custom_hooks = [dict(type='EMAHook',ema_type='ExpMomentumEMA',momentum=0.0001,update_buffers=True,strict_load=False,priority=49)
]val_evaluator = dict(type='mmdet.CocoMetric',proposal_nums=(100, 1, 10),ann_file=data_root + val_ann_file,metric='bbox')
test_evaluator = val_evaluatortrain_cfg = dict(type='EpochBasedTrainLoop',max_epochs=max_epochs,val_interval=save_checkpoint_intervals)
val_cfg = dict(type='ValLoop')
test_cfg = dict(type='TestLoop')