目标检测(Object Detection)是深度学习 CV 领域的一个核心研究领域和重要分支。纵观 2013 年到 2019 年,从最早的 R-CNN、Fast R-CNN 到后来的 YOLO v2、YOLO v3 再到今年的 M2Det,新模型层出不穷,性能也越来越好!本文将会对目标检测近几年的发展和相关论文做出一份系统介绍,总结一份超全的文献 paper 列表。
模型列表先一睹为快!(建议收藏)
这份目标检测超全的技术路线总结来自于 GitHub 上一个知名项目,作者是 Lee hoseong,项目地址是:
https://github.com/hoya012/deep_learning_object_detection
该技术路线横跨时间是 2014 年至 2019 年,上图总结了这期间目标检测所有重要的模型。图中标红的部分是作者认为比较重要,需要重点掌握的模型。当然每个人有都有各自的评价。
模型性能比较
FPS(速度)索引与硬件规格(如 CPU、GPU、RAM 等)有关,因此很难进行同等比较。解决方案是在具有相同规格的硬件上测量所有模型的性能,但这是非常困难和耗时的。比较结果如下:
、
下面举例对标红的重要模型进行介绍!
2014 年
R-CNN
Rich feature hierarchies for accurate object detection and semantic segmentation | Ross Girshick, Jeff Donahue, Trevor Darrell, Jitendra Malik | [CVPR' 14]
论文:
https://arxiv.org/pdf/1311.2524.pdf
代码 Caffe:
https://github.com/rbgirshick/rcnn
OverFeat
OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks | Pierre Sermanet, et al. | [ICLR' 14]
论文:
https://arxiv.org/pdf/1312.6229.pdf
代码 Torch:
https://github.com/sermanet/OverFeat
2015 年
Fast R-CNN
Fast R-CNN | Ross Girshick | [ICCV' 15]
论文:
https://arxiv.org/pdf/1504.08083.pdf
代码 caffe:
https://github.com/rbgirshick/fast-rcnn
Faster R-CNN
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks | Shaoqing Ren, et al. | [NIPS' 15]
论文:
代码 caffe:
https://github.com/rbgirshick/py-faster-rcnn
代码 tensorflow:
https://github.com/endernewton/tf-faster-rcnn
代码 pytorch:
https://github.com/jwyang/faster-rcnn.pytorch
2016 年
OHEM
Training Region-based Object Detectors with Online Hard Example Mining | Abhinav Shrivastava, et al. | [CVPR' 16]
论文:
https://arxiv.org/pdf/1604.03540.pdf
代码 caffe:
https://github.com/abhi2610/ohem
YOLO v1
You Only Look Once: Unified, Real-Time Object Detection | Joseph Redmon, et al. | [CVPR' 16]
论文:
https://arxiv.org/pdf/1506.02640.pdf
代码 c:
https://pjreddie.com/darknet/yolo/
SSD
Single Shot MultiBox Detector | Wei Liu, et al. | [ECCV' 16]
论文:
https://arxiv.org/pdf/1512.02325.pdf
代码 caffe:
https://github.com/weiliu89/caffe/tree/ssd
代码 tensorflow:
https://github.com/balancap/SSD-Tensorflow
代码 pytorch:
https://github.com/amdegroot/ssd.pytorch
R-FCN
Object Detection via Region-based Fully Convolutional Networks | Jifeng Dai, et al. | [NIPS' 16]
论文:
https://arxiv.org/pdf/1605.06409.pdf
代码 caffe:
https://github.com/daijifeng001/R-FCN
代码 caffe:
https://github.com/YuwenXiong/py-R-FCN
2017 年
YOLO v2
Better, Faster, Stronger | Joseph Redmon, Ali Farhadi | [CVPR' 17]
论文:
https://arxiv.org/pdf/1612.08242.pdf
代码 c:
https://pjreddie.com/darknet/yolo/
代码 caffe:
https://github.com/quhezheng/caffe_yolo_v2
代码 tensorflow:
https://github.com/nilboy/tensorflow-yolo
代码 tensorflow:
https://github.com/sualab/object-detection-yolov2
代码 pytorch:
https://github.com/longcw/yolo2-pytorch
FPN
Feature Pyramid Networks for Object Detection | Tsung-Yi Lin, et al. | [CVPR' 17]
论文:
代码 caffe:
RetinaNet
Focal Loss for Dense Object Detection | Tsung-Yi Lin, et al. | [ICCV' 17]
论文:
https://arxiv.org/pdf/1708.02002.pdf
代码 keras:
https://github.com/fizyr/keras-retinanet
代码 pytorch:
https://github.com/kuangliu/pytorch-retinanet
代码 mxnet:
https://github.com/unsky/RetinaNet
代码 tensorflow:
https://github.com/tensorflow/tpu/tree/master/models/official/retinanet
Mask R-CNN
Kaiming He, et al. | [ICCV' 17]
论文:
http://openaccess.thecvf.com/content_ICCV_2017/papers/He_Mask_R-CNN_ICCV_2017_paper.pdf
代码 caffe2:
https://github.com/facebookresearch/Detectron
代码 tensorflow:
https://github.com/matterport/Mask_RCNN
代码 tensorflow:
https://github.com/CharlesShang/FastMaskRCNN
代码 pytorch:
https://github.com/multimodallearning/pytorch-mask-rcnn
2018 年
YOLO v3
An Incremental Improvement | Joseph Redmon, Ali Farhadi | [arXiv' 18]
论文:
https://pjreddie.com/media/files/papers/YOLOv3.pdf
代码 c:
https://pjreddie.com/darknet/yolo/
代码 pytorch:
https://github.com/ayooshkathuria/pytorch-yolo-v3
代码 pytorch:
https://github.com/eriklindernoren/PyTorch-YOLOv3
代码 keras:
https://github.com/qqwweee/keras-yolo3
代码 tensorflow:
https://github.com/mystic123/tensorflow-yolo-v3
RefineDet
Single-Shot Refinement Neural Network for Object Detection | Shifeng Zhang, et al. | [CVPR' 18]
论文:
代码 caffe:
https://github.com/sfzhang15/RefineDet
代码 chainer:
https://github.com/fukatani/RefineDet_chainer
代码 pytorch:
https://github.com/lzx1413/PytorchSSD
2019 年
M2Det
A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network | Qijie Zhao, et al. | [AAAI' 19]
论文:
https://arxiv.org/pdf/1811.04533.pdf
参考文献
该项目的参考文献来自于论文《Deep Learning for Generic Object Detection: A Survey》
论文地址: