Skip to content

The repository not noly provides basic books and classic papers for learning AI, but also cutting-edge researchs.

Notifications You must be signed in to change notification settings

StarLab317/Learning-AI-All-In-One

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 

Repository files navigation

image

Learning AI All In One

梳理关于AI在Computer Vision/Robotics方向上的学习和应用: "AI is just a tool, but a powerful tool for almost anything."

[TOC]

1. 基础知识

课内必备的基础知识有助于更快速地上手

  • Python
  • Opencv
  • Linux系统操作
  • 线代
  • 概率论与数理统计
  • 数据挖掘

2. Machine Learning/Statistics Learning

如果对自己的数学有一定的自信,愿意从头理解底层算法而不仅仅是当一个调包侠,想要了解机器学习原理的同学,这绝对是你想要的学习路线

  • 李航-统计学习分析
    • 了解基础的有监督、无监督算法及其传统应用
  • NYU-Machine Learning
    • 强推这个NYU的Data Science项目,个人认为不仅比吴恩达Machine Learning视频更注重底层原理,而且比UCB的Machine Learning更关注实战与应用。
    • 全英教学,从统计和数学的角度去理解机器学习,最后回归到实践应用,解决实际问题。比如最后一个作业就是让你自己动手写一个神经网络的框架(Computational Gragh),并且帮你准备好了测试代码。绝大多数的算法工程师的面试的基础知识都能被这门课的slides给cover到。
  • NJU(Lamda)周志华-机器学习
    • 国内最经典的机器学习教程

3. DeepLearning

理解了Machine Learning/Statistics Learning的本质和算法后,就可以开始快速入门DeepLearning的经典算法和模型,此处要求对经典算法的核心思想、实现细节进行准确的把握

参考B站李沐的视频

3.1 分类

3.2 检测与识别

3.3 语义分割

4. Cutting-edge Research

强烈推荐李沐老师、同济子豪兄的前沿论文解读

4.1 Semi-Supervised Learning

4.2 3D Facial Expression Manipulation

4.3 3D Detection and Segmentation

4.4 BEV

4.5 3D Point

4.6 CLIP List

论文列表

1、 PromptDet: Towards Open-vocabulary Detection using Uncurated Images(ECCV2022)

2、 CLIP-Art: Contrastive Pre-training for Fine-Grained Art Classification(CVPR2021 workshops)

3、 CoOp: Learning to Prompt for Vision-Language Models(CVPR2021)

4、 DenseCLIP: Language-Guided Dense Prediction with Context-Aware Prompting(CVPR2022)

5、 Better Vision-Language Models with Feature Adapters(CVPR2021)

6、 ProposalCLIP: Unsupervised Open-Category Object Proposal Generation via Exploiting CLIP Cues(CVPR2022)

7、 Towards Open-vocabulary Scene Graph Generation with Prompt-based Finetuning

8、 StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery

9、 CRIS: CLIP-Driven Referring Image Segmentation

About

The repository not noly provides basic books and classic papers for learning AI, but also cutting-edge researchs.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published