The introduction

You just read theory, but you never learn? Can’t read it, and there’s no progress? If you have this problem, read this column. Based on actual projects, I will start with simple handwriting recognition, face detection and image classification and gradually learn advanced skills such as case segmentation, target detection and lane line detection. Learning is rewarding and fulfilling so you can keep going.

I have received too many private messages from people on Zhihu and the official account, looking for machine vision projects that can be operated in detail. For a variety of reasons, the response is always patchy. After more than half a year of collection and sorting, I decided to open this column, oriented by project practice, take you hand in hand from the basic handwriting recognition, image/video coding, gradually into facial expression recognition, lane detection, instance segmentation and other specific projects.

The items included in this column are not entirely original, but they are carefully vetted, citated, annotated and redacted. Such as handwriting recognition, face recognition, expression recognition, image segmentation, surface defect detection, lane line detection, vehicle passable area detection and other projects, provide complete guidance documents and open source code.

Continue to update, “practice” to promote learning.

directory

    • Theory of article

      • Machine Learning Algorithms and Python Implementation
      • Advanced Paper Study of Computer Vision

    • Introductory article

    • Advanced article

    • Practice guidelines

    • CV System Project

Theory of article

The projects covered in this column do not require in-depth knowledge of machine learning/image processing, but I have also written two columns, “Machine Learning Algorithms and Python Implementation” and “Advanced Papers in Computer Vision”. One is more algorithmic theory, and the other focuses on the cutting-edge paper results of computer vision, interpreting new methods and ideas.

Machine Learning Algorithms and Python Implementation

This column covers machine learning algorithm principles, in-depth analysis of decision trees, Bayesian algorithms, logistic regression, gradient descent, integrated learning, K-nearest Neighbor, support vector machines, etc. It also provides Python implementation sources, and shares practical guidance for Kaggle.

  • Column links blog.csdn.net/charmve/cat…

Advanced Paper Study of Computer Vision

Read computer vision papers, CVPR, ECCV, ICCV, etc., also have part of the project open source to Github, papers, source code, data sets.

Link: github.com/Charmve/Mir…

  • Column links charmve.blog.csdn.net/category_10…

Introductory article

  • Machine learning practical explanation (2) – | convolution neural network handwritten a convolutional neural network
  • Tutorial | deep learning + OpenCV, Python implementation real-time video target detection
  • CV of actual combat | using OpenCV panoramic image stitching
  • CV of actual combat | use OpenCV implement road vehicle counting
  • | this tutorial how to use the Docker, TensorFlow target detection API and OpenCV real-time target detection and video processing…
  • Keras realizes CNN: Handwritten digit recognition accuracy 99.6%
  • The application of machine learning field | logic return “Kaggle Titanic disaster”

Advanced article

  • Michael micro vision | with CenterNet of rotation target detection
  • Facial expression recognition based on CNN FER | classification of facial expression recognition research
  • Facial expression recognition FER | facial expression recognition paper reviews CVPR2019
  • Surface defect inspection summary data sets and the related project recommendation | making open source
  • Interpretation of ECCV 2020 GigaVision Challenge “Pedestrian and Vehicle Detection” and “Multi-target Tracking” winner proposal
  • ​Kaggle X-ray pneumonia test match second program parsing | CVPR 2020 Workshop
  • Processing bad data in facial expression recognition: an interpretation of CVPR 2020 and two TIPS
  • ECCV2020 Open source paper collection of image segmentation
  • BMVC2020 best paper | adaptive anti-aliasing convolutional neural network
  • Ross, He Kaiming et al. proposed the PointRend: rendering idea for image segmentation, which significantly improved the performance of Mask R-CNN
  • This video ‘eraser’ makes you disappear in seconds, leaving no hair on your head
  • CVPR2020 | glass detection in real situations, the interesting applications
  • Sliding Windows can also be used for instance segmentation. Chen Xinlei, He Kaiming et al proposed a new paradigm of image segmentation

Practice guidelines

  • Deep Learning Environment Configuration Guide! (Windows, Mac, Ubuntu)
  • 22 notes on refined diagrams, a prerequisite for deep learning
  • Why is your model so poor, and what are the techniques of deep learning tuning?
  • Does CNN really need downsampling (upsampling)?
  • Explore further: Why feature normalization/standardization?
  • Don’t you know about optimization algorithms in machine learning? Now I’ve summed it up for you
  • Machine learning paper reappears, here are five issues you need to pay attention to

CV System Project

  • Competition project | is based on the analysis of the emotional intelligence speech AI interactive entertainment system

  • Intelligent voice accompanying robot based on facial expression analysis

In the continuous update, if you encounter any problems in practice, welcome to communicate. If you reply “add group” in the public account, you can enter the Learning exchange group of Micro.



Follow public accountMai Micro AI InstituteLearn with more alchemists. And don’t forget to subscribe!

Column links: charmve.blog.csdn.net/article/det…