DL之FasterR-CNN:Faster R-CNN算法的简介(论文介绍)、架构详解、案例应用等配图集合之详细攻略
DL之FasterR-CNN:Faster R-CNN算法的簡介(論文介紹)、架構(gòu)詳解、案例應(yīng)用等配圖集合之詳細攻略
?
?
?
?
目錄
Faster R-CNN算法的簡介(論文介紹)
1、實驗結(jié)果
2、三者架構(gòu)對比——R-CNN、Fast?R-CNN、Faster R-CNN
Faster R-CNN算法的架構(gòu)詳解
1、Faster R-CNN
Faster R-CNN算法的案例應(yīng)用
Faster R-CNN思路結(jié)構(gòu)框圖
1、RPN網(wǎng)絡(luò)結(jié)構(gòu)
2、Anchor機制
?
?
?
?
?
?
相關(guān)文章
DL之R-CNN:R-CNN算法的簡介(論文介紹)、架構(gòu)詳解、案例應(yīng)用等配圖集合之詳細攻略
DL之FastR-CNN:Fast R-CNN算法的簡介(論文介紹)、架構(gòu)詳解、案例應(yīng)用等配圖集合之詳細攻略
DL之FasterR-CNN:Faster R-CNN算法的簡介(論文介紹)、架構(gòu)詳解、案例應(yīng)用等配圖集合之詳細攻略
DL之FasterR-CNN:Faster R-CNN算法的架構(gòu)詳解
Faster R-CNN算法的簡介(論文介紹)
? ? ??Faster R-CNN,顧名思義,相對R-CNN有非常大的提高!
Abstract
? ? ? State-of-the-art object detection networks depend on region proposal algorithms to hypothesize object locations. ?Advances like SPPnet [1] and Fast R-CNN [2] have reduced the running time of these detection networks, exposing region ?proposal computation as a bottleneck. In this work, we introduce a Region Proposal Network (RPN) that shares full-image ?convolutional features with the detection network, thus enabling nearly cost-free region proposals. An RPN is a fully convolutional ?network that simultaneously predicts object bounds and objectness scores at each position. The RPN is trained end-to-end to ?generate high-quality region proposals, which are used by Fast R-CNN for detection. We further merge RPN and Fast R-CNN ?into a single network by sharing their convolutional features—using the recently popular terminology of neural networks with ?“attention” mechanisms, the RPN component tells the unified network where to look. For the very deep VGG-16 model [3], ?our detection system has a frame rate of 5fps (including all steps) on a GPU, while achieving state-of-the-art object detection ?accuracy on PASCAL VOC 2007, 2012, and MS COCO datasets with only 300 proposals per image. In ILSVRC and COCO ?2015 competitions, Faster R-CNN and RPN are the foundations of the 1st-place winning entries in several tracks. Code has been ?made publicly available.
摘要
? ? ? 最先進的目標檢測網(wǎng)絡(luò)依賴于區(qū)域建議算法來假設(shè)目標位置。SPPnet[1]和Fast?R-CNN[2]等技術(shù)的進步,降低了檢測網(wǎng)絡(luò)的運行時間,暴露了區(qū)域提案計算的瓶頸。在這項工作中,我們引入了一個與檢測網(wǎng)絡(luò)共享全圖像卷積特性的區(qū)域建議網(wǎng)絡(luò)(RPN),從而實現(xiàn)了幾乎免費的區(qū)域建議。RPN是一個完全卷積的網(wǎng)絡(luò),它同時預(yù)測每個位置的對象邊界和對象得分。對RPN進行端到端訓(xùn)練,生成高質(zhì)量的區(qū)域建議,Fast R-CNN對其進行檢測。通過共享卷積特性,我們進一步將RPN和Fast R-CNN合并成一個單獨的網(wǎng)絡(luò)——使用最近流行的具有“注意”機制的神經(jīng)網(wǎng)絡(luò)術(shù)語,RPN組件告訴統(tǒng)一的網(wǎng)絡(luò)去哪里看。對于非常深的VGG-16型號[3],我們的檢測系統(tǒng)在GPU上的幀率為5fps(包括所有步驟),同時在PASCAL VOC 2007、2012和MS COCO數(shù)據(jù)集上實現(xiàn)了最先進的目標檢測精度,每張圖像只有300個提案。在ILSVRC和COCO 2015年的比賽中,Faster R-CNN和RPN是在多個賽道上獲得第一名的基礎(chǔ)。代碼已經(jīng)公開。
CONCLUSION ?
? ? ? We have presented RPNs for efficient and accurate ?region proposal generation. By sharing convolutional features with the down-stream detection network, the ?region proposal step is nearly cost-free. Our method ?enables a unified, deep-learning-based object detection ?system to run at near real-time frame rates. The ?learned RPN also improves region proposal quality ?and thus the overall object detection accuracy.
結(jié)論
? ? ? 為了高效、準確地生成區(qū)域建議,我們提出了一種新的區(qū)域建議生成方法。通過與下游檢測網(wǎng)絡(luò)共享卷積特性,區(qū)域建議步驟幾乎是免費的。我們的方法使一個統(tǒng)一的,基于深度學(xué)習(xí)的目標檢測系統(tǒng)運行在接近實時幀率。學(xué)習(xí)的RPN還提高了區(qū)域建議質(zhì)量,從而提高了總體目標檢測精度。
論文
Shaoqing Ren, KaimingHe, Ross Girshick, and Jian Sun.
Faster R-CNN: Towards real-time object detection with region proposal networks. NIPS, 2015
https://arxiv.org/abs/1506.01497v3
?
1、實驗結(jié)果
1、PASCAL VOC 2007
? ? ? ?Example detections using RPN proposals on PASCAL VOC 2007 test. 下圖為在PASCAL VOC 2007測試中,使用RPN進行目標檢測的結(jié)果。The proposed method detects objects in a wide range of scales and aspect ratios. 該方法檢測的目標對象,具有較寬的尺度和寬高比。
Detection results on PASCAL VOC 2007 test set
SS指采用選擇性搜索但沒有采用RPN的網(wǎng)絡(luò);unshared是指沒有共享特征的網(wǎng)絡(luò)。
RPN+VGG+shared能夠得到最好的結(jié)果!
?
?
2、PASCAL VOC 2012
Detection results on PASCAL VOC 2012 test set
RPN+VGG+shared能夠得到最好的結(jié)果!
測試的速度:VGG+SS+Fast R-CNN來說,每秒0.5幀,即處理一幀(幅圖像)大概需要2秒。
VGG+RPN+Fast R-CNN來說,處理一幀(幅圖像)大概需要0.2秒。
ZF網(wǎng)絡(luò)更快,每秒17幀(圖像),
?
3、R-CNN Test-Time Speed
整個比較來說,Faster R-CNN的提速是非常明顯的!而基于VGG的Fast R-CNN也只需要2.3秒左右即可!
?
?
?
?
2、三者架構(gòu)對比——R-CNN、Fast?R-CNN、Faster R-CNN
| R-CNN | Fast?R-CNN | Faster R-CNN |
?
?
?
Faster R-CNN算法的架構(gòu)詳解
DL之FasterR-CNN:Faster R-CNN算法的架構(gòu)詳解
1、Faster R-CNN
Faster R-CNN = Fast R-CNN + RPN (Region Proposal Networks)
- 不依賴于外部區(qū)域建議算法
- 單個前向傳播中進行目標檢測
“attention” mechanisms(注意力機制): the RPN component tells the unified network where to look
?
?
?
Faster R-CNN算法的案例應(yīng)用
后期更新……
?
?
?
Faster R-CNN思路結(jié)構(gòu)框圖
?
1、RPN網(wǎng)絡(luò)結(jié)構(gòu)
?
2、Anchor機制
?
?
?
?
參加文章
Faster RCNN:RPN,anchor,sliding windows
?
?
?
?
?
?
總結(jié)
以上是生活随笔為你收集整理的DL之FasterR-CNN:Faster R-CNN算法的简介(论文介绍)、架构详解、案例应用等配图集合之详细攻略的全部內(nèi)容,希望文章能夠幫你解決所遇到的問題。
- 上一篇: Py之playsound:playsou
- 下一篇: ML之MaL: 流形学习MaL的概念认知