目标检测算法盘点
目标检测算法在准确度(AP)和速度(speed)的对比:
论文中的知识点
评价指标AP(mAP)
- 最完整的检测模型评估指标mAP计算指南(附代码)_慕课手记:
边框回归(Bounding Box Regression)
IOU
非极大值抑制(non maximum suppression)
OHEM(Training Region-based Object Detectors with Online Hard Example Mining)
论文地址:https://arxiv.org/pdf/1604.03540.pdf
论文解读:
RPN(Region Proposal Network)
最新论文
CVPR2018 目标检测(object detection)算法总览
ECCV2018目标检测(object detection)算法总览
超越YOLOv3!普林斯顿大学提出:CornerNet-Lite,基于关键点的目标检测算法,已开源!
one-stage detectors
SSD(Single Shot MultiBox Detector) 论文地址:https://arxiv.org/pdf/1512.02325.pdf
论文解读:
YOLO(You only look once)
论文地址:
-
You Only Look Once: Unified, Real-Time Object Detection https://arxiv.org/pdf/1506.02640.pdf
-
YOLO9000: Better, Faster, Stronger https://arxiv.org/pdf/1612.08242.pdf
-
YOLOv3: An Incremental Improvement https://arxiv.org/pdf/1804.02767.pdf
代码地址:
- keras版本:https://github.com/httpqqwweee/keras-yolo3
- tensorflow版本:https://github.com/gliese581gg/YOLO_tensorflow
- PyTorch版本:https://github.com/liuyuemaicha/PyTorch-YOLOv3
论文解读:
YOLOv3: 训练自己的数据 - 点滴记录 - CSDN博客
目标检测之One-stage算法:YOLOv1-YOLOv3进化历程
keras源码解读:
探索 YOLO v3 实现细节 - 第4篇 数据和y_true
RetinaNet(Focal Loss for Dense Object Detection) 论文地址:https://arxiv.org/pdf/1708.02002.pdf
论文解读:
论文阅读: RetinaNet - Online Notes - CSDN博客
CornerNet 陈泰红:CornerNet:目标检测算法新思路
https://blog.csdn.net/u014380165/article/details/83032273
CenterNet OLDPAN:扔掉anchor!真正的CenterNet——Objects as Points论文解读
two-stage detectors
R-CNN(Region-based Convolutional Neural Networks) 论文地址:http://xxx.itp.ac.cn/pdf/1311.2524.pdf
Fast R-CNN 论文地址:http://xxx.itp.ac.cn/pdf/1504.08083.pdf
Faster R-CNN(Towards Real-Time Object Detection with Region Proposal Networks) 论文地址:http://xxx.itp.ac.cn/pdf/1506.01497.pdf
论文解读: R-CNN论文原理
Object Detection and Classification using R-CNNs
Mask-RCNN 论文地址:http://xxx.itp.ac.cn/pdf/1703.06870.pdf
论文解读:
Mask-RCNN技术解析 - 跟随技术的脚步-linolzhang的专栏 - CSDN博客
Mask RCNN笔记 - 生如蚁,美如神 - CSDN博客
TridentNet Naiyan Wang:TridentNet:处理目标检测中尺度变化新思路
其他
FPN(Feature Pyramid Networks for Object Detection) 论文地址:http://xxx.itp.ac.cn/pdf/1612.03144.pdf
论文解读: FPN(feature pyramid networks)算法讲解
FCN(Fully Convolutional Networks for Semantic Segmentation) 论文地址:https://arxiv.org/pdf/1411.4038.pdf
论文解读:https://link.zhihu.com/?target=https%3A//www.cnblogs.com/gujianhan/p/6030639.html
Rethinking ImageNet Pre-training - 何凯明大神的新作
论文地址:https://arxiv.org/pdf/1811.08883.pdf