# Object Detection Develop History-1 ==Two Stage Method== stage 1 = region proposal stage 2 = feature extract + bounding box R-CNN Fast R-CNN Faster R-CNN R-FCN --- ==R-CNN== 用傳統的computer vison 的selective search找region proposal 但是stage 2用CNN找FEATURE + BOUNDING BOX + SVM辨識目標種類 ![](https://i.imgur.com/0ghFz57.png) 缺點: 太慢 ---- ==Fast R-CNN== :::info 改進: 每個region proposal共用CNN來找feature&SVM換成Neural Network來加速 ::: ![](https://i.imgur.com/sC9bl3n.png) 缺點: 後面分類物體 + 給bounding box變快，前面的region proposal就顯得很慢 ---- ==Faster R-CNN== :::info 改進: region proposal也用CNN來加速 ::: ![](https://i.imgur.com/MxVyxjy.png) 缺點: 每個proposal區域的分類還是要單獨算一次 ---- ==R-FCN== :::info 改進: 整張影像的分類只要算一次 ::: ![](https://i.imgur.com/QJ1aAf9.png) ---- Two Stage Method Pros and Cons === Accuracy: High Speed: Slow(Relatively) 因此產生了不再分兩個階段的method ---> One Stage Method One Stage Method === Accuracy: Low Speed: fast(Relatively) ex. YOLO 將兩個階段的CNN合併成一個網路，同時選取proposal + classification 但是，還有一種加速two stage的method---> Light weighted Light weighted Two Stage === ![](https://i.imgur.com/U9ZYUEw.png) ex. PVANet --- Summary === ![](https://i.imgur.com/q3cqyob.png) :::warning YOLO已經有5代了 ::: ###### tags: `SAR` `Deep learning` `Object Detection` `review`