1. 針對對各技術特点分類的mind map:
圖片轉載 https://mp.weixin.qq.com/s/fiUUkT7hyJwXmAGQ1kMcqQ
2.一路照發展順序介紹了許多深度學習基本模塊:滑动窗口检测器、选择性搜索、边界框回归器、池化層、候选区域网络(RPN)、特征金字塔网络(FPN)、Focal Loss
Ref : 从RCNN到SSD目标检测算法盘点 https://www.jiqizhixin.com/articles/2018-04-27
3. 把2013-2020的deep learning object detection論文幾乎收錄的大礼包,除了論文連結有些還有官网源碼連結
還做了張發展roadmap並標註一些重大算法誔生的時間点!
Ref: https://github.com/hoya012/deep_learning_object_detection#2014
4. FAIR detectron:Facebook AI 研究院(FAIR)开源了 Detectron,业内集大成與一身的深度學習目标检测平台。並自創了一個"detectorn"名詞。不過必需用GPU跑,而且我用2GB內存的顯卡跑過還跑不動......
项目地址:https://github.com/facebookresearch/Detectron
在 FAIR 实验室,Detectron 目前已经支持很多研究项目的实现,包括:
-
Feature Pyramid Networks for Object Detection (https://arxiv.org/abs/1612.03144)
-
Mask R-CNN (https://arxiv.org/abs/1703.06870)
-
Detecting and Recognizing Human-Object Interactions (https://arxiv.org/abs/1704.07333)
-
Focal Loss for Dense Object Detection (https://arxiv.org/abs/1708.02002)
-
Non-local Neural Networks (https://arxiv.org/abs/1711.07971)
-
Learning to Segment Every Thing (https://arxiv.org/abs/1711.10370)
-
Data Distillation: Towards Omni-Supervised Learning (https://arxiv.org/abs/1712.04440)
Ref: 介紹detectron細節的博主 https://www.jiqizhixin.com/articles/2018-01-23-6