In this repository i tried to replicate a cool project by a japanese scientist who made a machine which had 100 % accuracy in defeating humans in the game of stone-paper and scissors
Alternatives To Stone_paper_scissor_defeator_using_opencv_keras
Project NameStarsDownloadsRepos Using ThisPackages Using ThisMost Recent CommitTotal ReleasesLatest ReleaseOpen IssuesLicenseLanguage
7 months ago26apache-2.0C++
vendor independent TinyML deep learning library, compiler and inference framework microcomputers and micro-controllers
10 months ago7C++
IO, scheduling, utils, machine learning... for Arduino
7 days ago2otherJupyter Notebook
This repository holds the Google Colabs for the EdX TinyML Specialization
Gestures Ml Js61
4 years ago2gpl-3.0JavaScript
[WIP] - Gesture recognition using hardware and Tensorflow.js
Rgb Neural Net60
5 years agogpl-3.0Jupyter Notebook
Physical visualisation of neural network learning using RGB leds, arduino and raspberry pi.
5 years ago1mitC++
Transcribe your voice by Google's Cloud Speech-to-Text API with esp32
Machine Learning For Physical Computing37
a year agoJavaScript
Repository for the "Machine Learning for Physical Computing" class at ITP, NYU
Arduino Library27
4 months ago1otherC++
This repository holds the Arduino Library for the EdX TinyML Specialization
Esp32 Autonomous Car25
7 months agomitPython
Autonomous car using ESP32.
Nindamani The Weed Removal Robot23
2 years agomitPython
Nindamani, the AI based mechanically weed removal robot
Alternatives To Stone_paper_scissor_defeator_using_opencv_keras
Select To Compare

Alternative Project Comparisons


In this repository i tried to replicate a cool project by a japanese scientist who made a machine which had 100 % accuracy in defeating humans in the game of stone-paper and scissors using convolutional neural networks and computer vision i have used opencv for computer vision and keras for CNNS link to video tutorial


  1. Python 3.x
  2. Tensorflow 1.5
  3. Keras
  4. OpenCV 3.4(for loading,resizing images)
  5. h5py(for saving trained model)
  6. pyttsx3
  7. A good grasp over convolutional neural networks. For online resources refer to standford cs231n, on coursera or cs231n by standford university
  8. A good CPU (preferably with a GPU).
  9. Time
  10. datetime
  11. Patience.... A lot of it.

Installing the requirements

  1. Start your terminal of cmd depending on your os.
  2. If you have a NVidia GPU then make sure you have the prerequisites for Tensorflow GPU installation (Refer to official site). Then use this commmand
pip install -r requirements_gpu.txt
  1. In case you do not have a GPU then use this command
pip install -r requirements_cpu.txt

Steps for making your own trainig data

Place the camera and don't move it, As soon as the camera starts, perform only one gesture at a time, the numberred images of this gesture will be stored in the root directory(you can modify the code append the path to which ever directory you want) Gather data for all the classes in the similar way

Training your own classifier

I used my own laptop for training puproses but you can use aws, google collab, azure ........ For training
1)Modify the path of the stone,paper and scissor folder in
After the model is trained you are ready to run it.

Running the trained model

1)Modify the path to the model file in

Congratulations you just made your very own MAN defeating machine.

Liked it

Give the repository a star if you really liked it.
if you have any doubts, you can comment them under my youtube video or you can post your doubts on my facebook page REACTOR SCIENCE
Thank you

Popular Machine Learning Projects
Popular Arduino Projects
Popular Machine Learning Categories
Related Searches

Get A Weekly Email With Trending Projects For These Categories
No Spam. Unsubscribe easily at any time.
Machine Learning
Convolutional Neural Networks
Computer Vision