Localization, Detection and Segmentation

2018-11-14

Image Classification이 아닌 다양한 Computer Vision Task 소개와 CNN Approach관점의 알고리즘 알아보겠습니다.

Computer Vision Task

  • Segmentation
  • Localization
  • Detection

목차

  • Image Classification이 아닌 Other Computer Vision Task들은 무엇이 있을까?
  • Segmentation, Localization, Detection 개념
  • Segmentation 종류
  • 아이디어 / 매카니즘(약간은 Case-study 느낌)
  • Single Object / Multi Object
  • Semantic Segmentation
    • Unpooling / Max Unpooling / Transpose Convolution
  • Classification + Localization
    • Aside : Human Pose Estimation
  • Object Detection
    • Sliding Window(Naive Approach / Region Proposal Approach)
    • R-CNN (Region Proposal + CNN) 아이디어 / 매카니즘 / 한계
    • Fast R-CNN
    • (SPPNet)
    • Faster R-CNN
    • YOLO / SSD
    • Aside : Object Detection + Captioning = Dense Captioning
  • Instance Segmentation
    • Mask R-CNN

표

마크다운으로 표 만들어주는 사이트(markdown table generator)

Reference

  • https://junjiwon1031.github.io/2017/09/08/Single-Shot-Multibox-Detector.html
  • https://sites.google.com/site/bimprinciple/in-the-news/dibleoning-euliyonghangaegchegeomchulr-cnnyolossd
  • https://m.blog.naver.com/PostView.nhn?blogId=sogangori&logNo=221007697796&proxyReferer=https%3A%2F%2Fwww.google.co.kr%2F
  • https://m.blog.naver.com/sogangori/221009464294

본 포스팅에 대한 잘못된 점, 의견, 기술적 토론은 언제나 환영입니다.