A webcam-based 3x3x3 rubik's cube solver written in Python 3 and OpenCV.

Overview

Qbr

Qbr, pronounced as Cuber, is a webcam-based 3x3x3 rubik's cube solver written in Python 3 and OpenCV.

  • 🌈 Accurate color detection
  • 🔍 Accurate 3x3x3 rubik's cube detection
  • 🔠 Multilingual

Solve mode

solve mode

Calibrate mode

Isn't the default color detection working out for you? Use the calibrate mode to let Qbr be familiar with your cube's color scheme. If your room has proper lighting then this will give you a 99.9% guarantee that your colors will be detected properly.

Simply follow the on-screen instructions and you're ready to go.

calibrate mode calibrate mode success

Table of Contents

Introduction

The idea to create this came personally to mind when I started solving rubik's cubes. There were already so many professional programmers around the world who created robots that solve a rubik's cube in an ETA of 5 seconds and since 2016 in 1 second (link). That inspired me to create my own. I started using images only and eventually switched to webcam.

Installation

$ git clone --depth 1 https://github.com/kkoomen/qbr.git
$ cd qbr
$ python3 -m venv env
$ source ./env/bin/activate
$ pip3 install -r requirements.txt

Usage

Make sure you run source ./env/bin/activate every time you want to run the program.

Run Qbr:

$ ./src/qbr.py

This opens a webcam interface with the following things:

The first 9-sticker display (upper left corner)

This is preview mode. This will update immediately and display how Qbr has detected the colors.

The second 9-sticker display (upper left corner)

This is the snapshot state. When pressing SPACE it will create a snapshot in order to show you what state it has saved. You can press SPACE as many times as you'd like if it has been detected wrong.

Amount of sides scanned (bottom left corner)

The bottom left corner shows the amount of sides scanned. This is so you know if you've scanned in all sides before pressing ESC.

Interface language (top right corner)

In the top right corner you can see the current interface language. If you want to change the interface language you can press l to cycle through them. Continue to press l until you've found the right language.

Default language is set to English.

Available languages are:

  • English
  • Hungarian
  • Deutsch
  • French
  • Dutch
  • 简体中文

Full 2D cube state visualization (bottom right corner)

This visualization represents the whole cube state that is being saved and can be used to confirm whether the whole cube state has been scanned successfully.

Calibrate mode

The default color scheme contains the most prominent colors for white, yellow, red, orange, blue and green. If this can't detect your cube its colors properly then you can use calibrate mode.

Press c to go into calibrate mode in order to let Qbr be familiar with your cube's color scheme. Simply follow the on-screen instructions and you're ready to go.

Note: Your calibrated settings are automatically saved after you've calibrated your cube successfully. The next time you start Qbr it will automatically load it.

Tip: If you've scanned wrong, simple go out of calibrate mode by pressing c and go back into calibrate by pressing c again.

Getting the solution

Qbr checks if you have filled in all 6 sides when pressing ESC. If so, it'll calculate a solution if you've scanned it correctly.

You should now see a solution (or an error if you did it wrong).

How to scan your cube properly?

There is a strict way of scanning in the cube. Qbr will detect the side automatically, but the way you rotate the cube during the time you're scanning it is crucial in order for Qbr to properly calculate a solution. Make sure to follow the steps below properly:

  • Start off with the green side facing the camera and white on top, green being away from you. Start by scanning in the green side at this point.
  • After you've scanned in the green side, rotate the cube 90 or -90 degrees horizontally. It doesn't matter if you go clockwise or counter-clockwise. Continue to do this for the green, blue, red and orange sides until you are back at the green side.
  • You should now be in the same position like you started, having green facing the camera and white on top. Rotate the cube forward 90 degrees, resulting in green at the bottom and white facing the camera. Start scanning in the white side.
  • After you've scanned the white side, turn the cube back to how you started, having green in front again and white on top. Now rotate the cube backwards 90 degrees, resulting in green on top and yellow facing the camera. Now you can scan in the last yellow side.

If you've done the steps above correctly, you should have a solution from Qbr.

Keybindings

  • SPACE for saving the current state

  • ESC quit

  • c toggle calibrate mode

  • l switch interface language

Paramaters

You can use -n or --normalize to also output the solution in a "human-readable" format.

For example:

  • R will be: Turn the right side a quarter turn away from you.
  • F2 will be: Turn the front face 180 degrees.

Example runs

$ ./qbr.py
Starting position:
front: green
top: white

Moves: 20
Solution: U2 R D2 L2 F2 L U2 L F' U L U R2 B2 U' F2 D2 R2 D2 R2
$ ./qbr.py -n
Starting position:
front: green
top: white

Moves: 20
Solution: B2 U2 F' R U D' L' B' U L F U F2 R2 F2 D' F2 D R2 D2
1. Turn the back side 180 degrees.
2. Turn the top layer 180 degrees.
3. Turn the front side a quarter turn to the left.
4. Turn the right side a quarter turn away from you.
5. Turn the top layer a quarter turn to the left.
6. Turn the bottom layer a quarter turn to the left.
7. Turn the left side a quarter turn away from you.
8. Turn the back side a quarter turn to the right.
9. Turn the top layer a quarter turn to the left.
10. Turn the left side a quarter turn towards you.
11. Turn the front side a quarter turn to the right.
12. Turn the top layer a quarter turn to the left.
13. Turn the front side 180 degrees.
14. Turn the right side 180 degrees.
15. Turn the front side 180 degrees.
16. Turn the bottom layer a quarter turn to the left.
17. Turn the front side 180 degrees.
18. Turn the bottom layer a quarter turn to the right.
19. Turn the right side 180 degrees.
20. Turn the bottom layer 180 degrees.

Inspirational sources

Special thanks to HaginCodes for the main inspiration on how to improve my color detection.

https://github.com/HaginCodes/3x3x3-Rubiks-Cube-Solver

http://programmablebrick.blogspot.com/2017/02/rubiks-cube-tracker-using-opencv.html

https://gist.github.com/flyboy74/2cc3097f784c8c236a1a85278f08cddd

https://github.com/dwalton76/rubiks-color-resolver

License

Qbr is licensed under the MIT License.

Owner
Kim 金可明
Vim enthusiast; polyglot programmer; fullstack software engineer; QA engineer
Kim 金可明
Detect the mathematical formula from the given picture and the same formula is extracted and converted into the latex code

Mathematical formulae extractor The goal of this project is to create a learning based system that takes an image of a math formula and returns corres

6 May 22, 2022
An OCR evaluation tool

dinglehopper dinglehopper is an OCR evaluation tool and reads ALTO, PAGE and text files. It compares a ground truth (GT) document page with a OCR resu

QURATOR-SPK 40 Dec 20, 2022
An application of high resolution GANs to dewarp images of perturbed documents

Docuwarp This project is focused on dewarping document images through the usage of pix2pixHD, a GAN that is useful for general image to image translat

Thomas Huang 97 Dec 25, 2022
Natural language detection

Detect the language of text. What’s so cool about franc? franc can support more languages(†) than any other library franc is packaged with support for

Titus 3.8k Jan 02, 2023
Image augmentation for machine learning experiments.

imgaug This python library helps you with augmenting images for your machine learning projects. It converts a set of input images into a new, much lar

Alexander Jung 13.2k Jan 02, 2023
基于openpose和图像分类的手语识别项目

手语识别 0、使用到的模型 (1). openpose,作者:CMU-Perceptual-Computing-Lab https://github.com/CMU-Perceptual-Computing-Lab/openpose (2). 图像分类classification,作者:Bubbl

20 Dec 15, 2022
MXNet OCR implementation. Including text recognition and detection.

insightocr Text Recognition Accuracy on Chinese dataset by caffe-ocr Network LSTM 4x1 Pooling Gray Test Acc SimpleNet N Y Y 99.37% SE-ResNet34 N Y Y 9

Deep Insight 99 Nov 01, 2022
RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition

RepMLP RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition Released the code of RepMLP together with an example o

260 Jan 03, 2023
Rubik's Cube in pygame with OpenGL

Rubik Rubik's Cube in pygame with OpenGL The script show on the screen a Rubik Cube buit with OpenGL. Then I have also implemented all the possible mo

Gabro 2 Apr 15, 2022
Code for the head detector (HeadHunter) proposed in our CVPR 2021 paper Tracking Pedestrian Heads in Dense Crowd.

Head Detector Code for the head detector (HeadHunter) proposed in our CVPR 2021 paper Tracking Pedestrian Heads in Dense Crowd. The head_detection mod

Ramana Subramanyam 76 Dec 06, 2022
A community-supported supercharged version of paperless: scan, index and archive all your physical documents

Paperless-ngx Paperless-ngx is a document management system that transforms your physical documents into a searchable online archive so you can keep,

5.2k Jan 04, 2023
Unofficial implementation of "TableNet: Deep Learning model for end-to-end Table detection and Tabular data extraction from Scanned Document Images"

TableNet Unofficial implementation of ICDAR 2019 paper : TableNet: Deep Learning model for end-to-end Table detection and Tabular data extraction from

Jainam Shah 243 Dec 30, 2022
👄 The most accurate natural language detection library for Java and the JVM, suitable for long and short text alike

Quick Info this library tries to solve language detection of very short words and phrases, even shorter than tweets makes use of both statistical and

Peter M. Stahl 532 Dec 28, 2022
A buffered and threaded wrapper for the OpenCV VideoCapture object. Can speed up video decoding significantly. Supports

A buffered and threaded wrapper for the OpenCV VideoCapture object. Can speed up video decoding significantly. Supports "with"-syntax.

Patrice Matz 0 Oct 30, 2021
This is a implementation of CRAFT OCR method

This is a implementation of CRAFT OCR method

Esaka 0 Nov 01, 2021
https://arxiv.org/abs/1904.01941

Character-Region-Awareness-for-Text-Detection- https://arxiv.org/abs/1904.01941 Train You can train SynthText data use python source/train_SynthText.p

DayDayUp 120 Dec 28, 2022
This is a GUI for scrapping PDFs with the help of optical character recognition making easier than ever to scrape PDFs.

pdf-scraper-with-ocr With this tool I am aiming to facilitate the work of those who need to scrape PDFs either by hand or using tools that doesn't imp

Jacobo José Guijarro Villalba 75 Oct 21, 2022
Application that instantly translates sign-language to letters.

Sign Language Translator Project Description The main purpose of project is translating sign-language to letters. In accordance with this purpose we d

3 Sep 29, 2022
SemTorch

SemTorch This repository contains different deep learning architectures definitions that can be applied to image segmentation. All the architectures a

David Lacalle Castillo 154 Dec 07, 2022
document image degradation

ocrodeg The ocrodeg package is a small Python library implementing document image degradation for data augmentation for handwriting recognition and OC

NVIDIA Research Projects 134 Nov 18, 2022