This is a computer vision based implementation of the popular childhood game 'Hand Cricket/Odd or Even' in python

Overview

Hand Cricket

Table of Content

Overview

This is a computer vision based implementation of the popular childhood game 'Hand Cricket/Odd or Even' in python. Behind the game is a CNN model that is trained to identify hand sign for numbers 0,1,2,3,4,5 & 6. For those who have never played this game, the rules are explained below.

The Game in action

hand-cricket.mov

Installation

  • You need Python (3.6) & git (to clone this repo)
  • git clone [email protected]:abhinavnayak11/Hand-Cricket.git . : Clone this repo
  • cd path/to/Hand-Cricket : cd into the project folder
  • conda env create -f environment.yml : Create a virtual env with all the dependencies
  • conda activate comp-vision : activate the virtual env
  • python src/hand-cricket.py : Run the script

Game rules

Hand signs

  • You can play numbers 0, 1, 2, 3, 4, 5, 6. Their hand sign are shown here

Toss

  • You can choose either odd or even (say you choose odd)
  • Both the players play a number (say players play 3 & 6). Add those numbers (3+6=9).
  • Check if the sum is odd or even. (9 is odd)
  • If the result is same as what you have chosen, you have won the toss, else you have lost. (9 is odd, you chose odd, hence you win)

The Game

  • The person who wins the toss is the batsman, the other player is the bowler. (In the next version of the game, the toss winner will be allowed to chose batting/bowling)
  • Scoring Runs:
    • Both players play a number.
    • The batsman's number is added to his score only when the numbers are different.
    • There is special power given to 0. If batsman plays 0 and bowler plays any number but 0, bowler's number is added to batsman's score
  • Getting out:
    • Batsman gets out when both the players play the same number. Even if both the numbers are 0.
  • Winning/Losing:
    • After both the players have finished their innings, the person scoring more runs wins the game

Game code : hand-cricket.py


Project Details

  1. Data Collection :
    • After failing to find a suitable dataset, I created my own dataset using my phone camera.
    • The dataset contains a total of 1848 images. To ensure generality (i.e prevent overfitting to one type of hand in one type of environment) images were taken with 4 persons, in 6 different lighting conditions, in 3 different background.
    • Sample of images post augmentations are shown below, images
    • Data can be found uploaded at : github | kaggle. Data collection code : collect-data.py
  2. Data preprocessing :
    • A Pytorch dataset was created to handle the preprocessing of the image dataset (code : dataset.py).
    • Images were augmented before training. Following augmentations were used : Random Rotation, Random Horizontal Flip and Normalization. All the images were resized to (128x128).
    • Images were divided into training and validation set. Training set was used to train the model, whereas validation set helped validate the model performance.
  3. Model training :
    • Different pretrained models(resent18, densenet121 etc, which are pre-trained on the ImageNet dataset) from pytorch library were used to train on this dataset. Except the last 2 layers, all the layers were frozen and then trained. With this the pre-trained model helps extracting useful features and the last 2 layers will be fine-tuned to my dataset.
    • Learning rate for training the model was chosen with trial and error. For each model, learning rate was different.
    • Of all the models trained, densnet121 performed the best, with a validation accuracy of 0.994.
    • Training the model : train.py, engine.py, training-notebook

Future Scope

  • Although, this was a fun application, the dataset can be used in applications like sign language recognition.


License: MIT

Owner
Abhinav R Nayak
Aspiring data scientist
Abhinav R Nayak
ILVR: Conditioning Method for Denoising Diffusion Probabilistic Models (ICCV 2021 Oral)

ILVR + ADM This is the implementation of ILVR: Conditioning Method for Denoising Diffusion Probabilistic Models (ICCV 2021 Oral). This repository is h

Jooyoung Choi 225 Dec 28, 2022
git《Pseudo-ISP: Learning Pseudo In-camera Signal Processing Pipeline from A Color Image Denoiser》(2021) GitHub: [fig5]

Pseudo-ISP: Learning Pseudo In-camera Signal Processing Pipeline from A Color Image Denoiser Abstract The success of deep denoisers on real-world colo

Yue Cao 51 Nov 22, 2022
Learning Generative Models of Textured 3D Meshes from Real-World Images, ICCV 2021

Learning Generative Models of Textured 3D Meshes from Real-World Images This is the reference implementation of "Learning Generative Models of Texture

Dario Pavllo 115 Jan 07, 2023
My personal Home Assistant configuration.

About This is my personal Home Assistant configuration. My guiding princile is to have full local control of all my devices. I intend everything to ru

Chris Turra 13 Jun 07, 2022
This repo tries to recognize faces in the dataset you created

YÜZ TANIMA SİSTEMİ Bu repo oluşturacağınız yüz verisetlerini tanımaya çalışan ma

Mehdi KOŞACA 2 Dec 30, 2021
Swin-Transformer is basically a hierarchical Transformer whose representation is computed with shifted windows.

Swin-Transformer Swin-Transformer is basically a hierarchical Transformer whose representation is computed with shifted windows. For more details, ple

旷视天元 MegEngine 9 Mar 14, 2022
This is an official implementation for "PlaneRecNet".

PlaneRecNet This is an official implementation for PlaneRecNet: A multi-task convolutional neural network provides instance segmentation for piece-wis

yaxu 50 Nov 17, 2022
The codebase for our paper "Generative Occupancy Fields for 3D Surface-Aware Image Synthesis" (NeurIPS 2021)

Generative Occupancy Fields for 3D Surface-Aware Image Synthesis (NeurIPS 2021) Project Page | Paper Xudong Xu, Xingang Pan, Dahua Lin and Bo Dai GOF

xuxudong 97 Nov 10, 2022
Adversarial examples to the new ConvNeXt architecture

Adversarial examples to the new ConvNeXt architecture To get adversarial examples to the ConvNeXt architecture, run the Colab: https://github.com/stan

Stanislav Fort 19 Sep 18, 2022
NAACL2021 - COIL Contextualized Lexical Retriever

COIL Repo for our NAACL paper, COIL: Revisit Exact Lexical Match in Information Retrieval with Contextualized Inverted List. The code covers learning

Luyu Gao 108 Dec 31, 2022
FAMIE is a comprehensive and efficient active learning (AL) toolkit for multilingual information extraction (IE)

FAMIE: A Fast Active Learning Framework for Multilingual Information Extraction

18 Sep 01, 2022
[ICRA 2022] CaTGrasp: Learning Category-Level Task-Relevant Grasping in Clutter from Simulation

This is the official implementation of our paper: Bowen Wen, Wenzhao Lian, Kostas Bekris, and Stefan Schaal. "CaTGrasp: Learning Category-Level Task-R

Bowen Wen 199 Jan 04, 2023
Deep Learning Visuals contains 215 unique images divided in 23 categories

Deep Learning Visuals contains 215 unique images divided in 23 categories (some images may appear in more than one category). All the images were originally published in my book "Deep Learning with P

Daniel Voigt Godoy 1.3k Dec 28, 2022
Chinese clinical named entity recognition using pre-trained BERT model

Chinese clinical named entity recognition (CNER) using pre-trained BERT model Introduction Code for paper Chinese clinical named entity recognition wi

Xiangyang Li 109 Dec 14, 2022
The dataset of tweets pulling from Twitters with keyword: Hydroxychloroquine, location: US, Time: 2020

HCQ_Tweet_Dataset: FREE to Download. Keywords: HCQ, hydroxychloroquine, tweet, twitter, COVID-19 This dataset is associated with the paper "Understand

2 Mar 16, 2022
Code for Massive-scale Decoding for Text Generation using Lattices

Massive-scale Decoding for Text Generation using Lattices Jiacheng Xu, Greg Durrett TL;DR: a new search algorithm to construct lattices encoding many

Jiacheng Xu 37 Dec 18, 2022
Collection of tasks for fast prototyping, baselining, finetuning and solving problems with deep learning.

Collection of tasks for fast prototyping, baselining, finetuning and solving problems with deep learning Installation

Pytorch Lightning 1.6k Jan 08, 2023
COVID-VIT: Classification of Covid-19 from CT chest images based on vision transformer models

COVID-ViT COVID-VIT: Classification of Covid-19 from CT chest images based on vision transformer models This code is to response to te MIA-COV19 compe

17 Dec 30, 2022
LinkNet - This repository contains our Torch7 implementation of the network developed by us at e-Lab.

LinkNet This repository contains our Torch7 implementation of the network developed by us at e-Lab. You can go to our blogpost or read the article Lin

e-Lab 158 Nov 11, 2022
Codes to pre-train T5 (Text-to-Text Transfer Transformer) models pre-trained on Japanese web texts

t5-japanese Codes to pre-train T5 (Text-to-Text Transfer Transformer) models pre-trained on Japanese web texts. The following is a list of models that

Kimio Kuramitsu 1 Dec 13, 2021