一、绪论——Lecture 1: Introduction-Computer vision overview Historical context Course logistics——Description
说明
课程链接:http://cs231n.stanford.edu/schedule.html
参考笔记:https://zhuanlan.zhihu.com/p/21930884
b站视频:https://www.bilibili.com/video/BV1Qt4y187h8?p=1&vd_source=3f409e335d99edd58fc22f4c59f6ae9e
一、Description
Lecture 1: Introduction-Computer vision overview Historical context Course logistics(计算机视觉综述、历史背景、课程逻辑)
A brief history of computer vision & deep learning
psychology——心理学
biology——生物学
speech recognition——语音识别
有这么快吗
反向传播
Hiton,之前学胶囊网络(capsule network)时知道他很厉害
杨立昆教授
- Applied backprop algorithm to a Neocognitron-like architecture
将反向传播算法应用于类新认知电子架构 - Learned to recognize handwritten digits
学会了识别手写数字 - Was deployed in a commercial system by NEC, processed handwritten checks
被NEC部署在一个商业系统中,处理手写支票 - Very similar to our modern convolutional networks!
非常类似于我们现代的卷积网络!
AlexNet: Deep Learning Goes Mainstream
AlexNet:深度学习成为主流
The AI winter thawed, deep learning revolution arrived
人工智能的寒冬解冻,深度学习革命到来
2012 to Present: Deep Learning Explosion
2012年至今:深度学习大爆发
2012 to Present: Deep Learning is Everywhere
AI的爆炸性增长和影响
CS231n overview
There are many visual recognition problems that are related to image classification, such as object detection, image captioning, image segmentation, visual question answering, visual instruction navigation, video understanding, etc
与图像分类相关的视觉识别问题有很多,如目标检测、图像字幕、图像分割、视觉答题、视觉指令导航、视频理解等。
分类:没有空间范围
语义分割:没有目标,只是像素
目标检测:多目标
实例分割:多目标
著名顶会:CVPR、ICCV
本文来自博客园,作者:JaxonYe,转载请注明原文链接:https://www.cnblogs.com/yechangxin/articles/16524860.html
侵权必究