目标检测“Focal Loss for Dense Object Detection”
2017-09-22 17:04
811 查看
基于RCNN的两步检测器精度高,但速度慢,单步检测器如YOLO,SSD速度快,但精度有所降低。作者认为正负样本不平衡是导致单步检测器精度降低的主要愿意,论文对交叉熵损失函数进行改进,降低易分类样本的权值,即Focal loss,使用RetinaNet验证Focal loss。RetinaNet可以超过两步检测器的精度,且速度跟单步检测器差不多。
Focal Loss是动态缩放的交叉熵损失函数,随着正确类的置信度上升缩放因子下降,如下图所示:
损失函数为:
当样本分错时,pt很小,调节因子接近1,损失函数不受影响。当样本很容易分时,pt很大,因子趋于0,这样易分样本的权值下降了。参数γ决定了权值下降的速率,文中选取γ=2
RetinaNet网络结构,特征提取部分采用FPN,Anchors的分配类似RPN,改进后用于多类检测,FPN后有分类子网络和box回归子网络:
实验结果
Focal Loss是动态缩放的交叉熵损失函数,随着正确类的置信度上升缩放因子下降,如下图所示:
损失函数为:
当样本分错时,pt很小,调节因子接近1,损失函数不受影响。当样本很容易分时,pt很大,因子趋于0,这样易分样本的权值下降了。参数γ决定了权值下降的速率,文中选取γ=2
RetinaNet网络结构,特征提取部分采用FPN,Anchors的分配类似RPN,改进后用于多类检测,FPN后有分类子网络和box回归子网络:
实验结果
相关文章推荐
- 目标检测--Focal Loss for Dense Object Detection
- Focal Loss for Dense Object Detection
- Focal Loss for Dense Object Detection 论文阅读
- 《Focal Loss for Dense Object Detection》论文阅读笔记
- Focal Loss for Dense Object Detection
- [论文笔记]Focal Loss for Dense Object Detection
- 论文阅读-《Focal Loss for Dense Object Detection》
- Focal Loss for Dense Object Detection
- Focal Loss for Dense Object Detection
- Focal Loss for Dense Object Detection论文阅读
- Focal Loss for Dense Object Detection
- Focal Loss for Dense Object Detection
- 论文解读Focal Loss for Dense Object Detection
- [论文阅读] Focal Loss for Dense Object Detection
- Focal Loss for Dense Object Detection
- 目标检测“Perceptual Generative Adversarial Networks for Small Object Detection”
- 目标检测--Enhancement of SSD by concatenating feature maps for object detection
- [目标检测] RON-Reverse Connection with Objectness Prior Networks for Object Detection
- 《Focal Loss for Object Detection》阅读笔记
- 目标检测--RON: Reverse Connection with Objectness Prior Networks for Object Detection