31 lines
2.0 KiB
Markdown
31 lines
2.0 KiB
Markdown
|
# Loss系列
|
|||
|
|
|
|||
|
|
1. SlideLoss,EMASlideLoss.(可动态调节正负样本的系数,让模型更加注重难分类,错误分类的样本上)[Yolo-Face V2](https://github.com/Krasjet-Yu/YOLO-FaceV2/blob/master/utils/loss.py)
|
|||
|
|
|
|||
|
|
在ultralytics/utils/loss.py中的class v8DetectionLoss进行设定.(支持v8-detect、v8-seg、v8-pose、v10)
|
|||
|
|
EMASlideLoss具体思想可以参考：https://www.bilibili.com/video/BV1W14y1i79U/?vd_source=c8452371e7ca510979593165c8d7ac27
|
|||
|
|
|
|||
|
|
2. FocalLoss,VarifocalLoss,QualityfocalLoss(支持v8-detect、v8-seg、v8-pose、v8-obb、v10)
|
|||
|
|
|
|||
|
|
项目视频百度云链接-20240111版本更新说明.
|
|||
|
|
|
|||
|
|
3. Normalized Gaussian Wasserstein Distance(支持v8-detect、v8-seg、v8-pose、v10)[论文链接](https://arxiv.org/abs/2110.13389)
|
|||
|
|
|
|||
|
|
在Loss中使用:
|
|||
|
|
在ultralytics/utils/loss.py中的BboxLoss class中的__init__函数里面设置self.nwd_loss为True.
|
|||
|
|
比例系数调整self.iou_ratio, self.iou_ratio代表iou的占比,(1-self.iou_ratio)为代表nwd的占比.
|
|||
|
|
在TAL标签分配中使用:
|
|||
|
|
在ultralytics/utils/tal.py中的def iou_calculation函数中进行更换即可.
|
|||
|
|
以上这两可以配合使用,也可以单独使用.
|
|||
|
|
|
|||
|
|
4. 定位损失系列(支持v8-detect、v8-seg、v8-pose、v10)
|
|||
|
|
|
|||
|
|
1. IoU,GIoU,DIoU,CIoU,EIoU,SIoU,MPDIoU,ShapeIoU.
|
|||
|
|
2. Inner-IoU,Inner-GIoU,Inner-DIoU,Inner-CIoU,Inner-EIoU,Inner-SIoU,Inner-ShapeIoU,Inner-MPDIoU.
|
|||
|
|
3. Focaler-IoU系列(IoU,GIoU,DIoU,CIoU,EIoU,SIoU,WIoU,MPDIoU,ShapeIoU)
|
|||
|
|
4. Powerful-IoU,Powerful-IoUV2,Inner-Powerful-IoU,Inner-Powerful-IoUV2,Focaler-Powerful-IoU,Focaler-Powerful-IoUV2[论文链接](https://www.sciencedirect.com/science/article/abs/pii/S0893608023006640)
|
|||
|
|
项目视频百度云链接-定位损失系列更换说明
|
|||
|
|
|
|||
|
|
1. Wise-IoU(v1,v2,v3)系列(IoU,WIoU,EIoU,GIoU,DIoU,CIoU,SIoU,MPDIoU,ShapeIoU,Powerful-IoU,Powerful-IoUV2).
|
|||
|
|
2. Inner-Wise-IoU(v1,v2,v3)系列(IoU,WIoU,EIoU,GIoU,DIoU,CIoU,SIoU,MPDIoU,ShapeIoU,Powerful-IoU,Powerful-IoUV2).
|
|||
|
|
项目视频百度云链接-20240111版本更新说明
|