SMART SOLUTIONS TO DEVELOP CRICKET SKILLS

​

2021-172

TEAM MEMBERS

INTRODUCTION

• Cricket is played all over the world, regardless of age and gender.

​

• Do all players get enough practice?

​

• There are lot of people with passion towards cricket but due to financial issues and unavailability of coaches they are unable afford proper cricket coaching.

​

460 million

STATS

40%

0

World Cricket Playing Population

​

​

No of people who practice

No of application to practice cricket

• How to capture the body movement?
• What are the ways of capturing user's body movement ?
• How to extract the coordinates from the image and videos?

RESEARCH QUESTIONS

ASSISTING USER

BATTING

OBJECTIVES

STANCE

​

WICKET-TAKING PERCENTAGE

Improve their batting shots

Guide the user to learn the batting stance

Provide suggestions to improve their cricket skills

Increase their wicket taking ability

DEPLOYMENTS AND

ACHIEVEMENTS

• Achievements
• Selected as one of the 25 teams to represent SLIIT at NBQSA
• Recognized by local Cricket Clubs - CCC
• Submitted research paper at ICAC Conference

• Deployments

SOLUTION

R.VITHURSON | IT18142556

BSc Hons information Technology specialization in Software Engineering

​

R. VITHURSON | IT18142556 | PROJECT ID - 2021-172

INTRODUCTION

• Detecting Human Pose in User Input Image.
• Identifying pose similarity between images .
• Human interactable mobile application to cricket related problems.

R.VITHURSON | IT18142556 | PROJECT ID - 2021-172

• How to detect the human pose?
• How to identify the pose similarity?
• How to extract the coordinates?
• How to get a high accuracy result ?

RESEARCH QUESTION

R.VITHURSON | IT18142556 | PROJECT ID - 2021-172

Main Objectives

• Detecting Human pose from user input image.
• Finding pose similarity.

Sub-Objectives

• Detecting the human pose and extracting the coordinates
• Using coordinates finding the pose similarity.
• Providing extracted data to trained model.

OBJECTIVES

R.VITHURSON | IT18142556 | PROJECT ID - 2021-172

METHODOLOGY

• Human pose detection tried with various technologies such as TFpose ,OpenCV, Google video API ,Media pipe
• After comparing all the results we found Media Pipe gave the High Accuracy.
• To obtain image pose similarity we used python libraries extract coordinates.

​

R.VITHURSON | IT18142556 | PROJECT ID - 2021-172

METHODOLOGY

R.VITHURSON | IT18142556 | PROJECT ID - 2021-172

• To extract human pose in media pipe they used their own BlazeFace model .

• The media pipe library extracts 33 body key points .(X,Y coordinates) but we use only 15 keypoints .

COMPONENT OVERVIEW

R.VITHURSON | IT18142556 | PROJECT ID - 2021-172

HUMAN POSE DETECTION

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

TECHNOLOGIES

App Development

Manipulate data

Transfer Data

Cricket Shot Detection

Storage

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

REFERENCES

[1] Foysal, Md. Ferdouse & Islam, Mohammad & Karim, Asif & Neehal, Nafis. (2018). Shot-Net: A Convolutional Neural Network for Classifying Different Cricket Shots.

[2]A. Z. M. E. Chowdhury, M. S. Rahim and M. A. U. Rahman, "Application of computer vision in Cricket: Foot overstep no-ball detection," 2016 3rd International Conference on Electrical Engineering and Information Communication Technology (ICEEICT), Dhaka, 2016, pp. 1-5, doi: 10.1109/CEEICT.2016.7873086

​

D.MITHELAN | IT18125344

BSc Hons information Technology specialization in Software Engineering

​

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

INTRODUCTION

• Classify the cricket shot performed by the user’s input.
• Similarity estimation for user’s input with an International player.
• Mobile Application for all the people who have an interest on cricket.

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

• How to capture the body coordinates?
• How to classify the cricket shot?
• How to get better accuracy than existing solutions?

RESEARCH QUESTION

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

Main Objectives

• Identifying the type of cricket shot is played.
• Similarity estimation of the cricket shot with an international player.

Sub-Objectives

• Extract the coordinates of each shot.
• Classifying the coordinates of the cricket shot using decision tree algorithms.
• Obtaining the body coordinates from the user.
• Assist the user with stats

OBJECTIVES

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

Six types of shot

METHODOLOGY

• First tried with TF-pose estimation.
• By using Mediapipe Pose estimation, we can get the coordinates.
• Random Forest Algorithm to classify the

cricket shots.

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

METHODOLOGY

• Initially, collected 900 images for Cut,Straight and Cover Drive.
• Then ,gathered 3600 images of dataset from previous

authors [1]

• 2209/3600 images were extracted from the dataset.
• Performed same approach for 2209 images, carried out by the existing solution [1]
• [1] system got 80 % accuracy for 3600 images using CNN and for 2209 images - 81 %

​

​

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

COMPONENT OVERVIEW

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

RESULTS & DISCUSSIONS

Shotnet [1]

Cricket4u

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

For 2209 images extracted

CLASSIFICATION

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

TECHNOLOGIES

App Development

Manipulate data

Transfer Data

Cricket Shot Detection

ML Library

Storage

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

REFERENCES

[1] Foysal, Md. Ferdouse & Islam, Mohammad & Karim, Asif & Neehal, Nafis. (2018). Shot-Net: A Convolutional Neural Network for Classifying Different Cricket Shots.

[2] M. Zeeshan Khan, M. A. Hassan, A. Farooq and M. U. Ghanni Khan, "Deep CNN Based Data-Driven Recognition of Cricket Batting Shots," 2018 International Conference on Applied and Engineering Mathematics (ICAEM), Taxila, Pakistan, 2018, pp. 67-71, doi: 10.1109/ICAEM.2018.8536277.

​

​

D. MITHELAN | IT18125344 | PROJECT ID - 2021-172

A.Manoj Karthik | IT18190762

BSc Hons information Technology specialization in Information technology

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

INTRODUCTION

•Identify the bowlers bowling length.

•Identifying the Ball

•Identifying the speed of the ball.

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

​

RESEARCH QUESTION

•How to identify the ball and bowlers’ length ?

•How to identify the bowlers bowling speed ?

•How to deliver the tracked records ?

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

​

OBJECTIVES

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

​

Main Objectives​

• To identify a bowler’s bowling length and the ball's speed.​
• Providing a detailed report to user about their bowling performance.​

​

Sub Objectives ​

• Getting the video file as input from user​
• Identifying the pitching location using darknet YOLOv4 Algorithm ​

​

METHODOLOGY

• Getting the input from the user as a video file.

​

• Identifying the ball, pitching position of the ball and batsman’s position using Yolov4.

​

• Pass the ball coordinates and batsman’s coordinates to predict the wicket taking percentage

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

​

COMPONENT OVERVIEW

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

​

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

​

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

​

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

​

TECHNOLOGIES

App Development

Manipulate data

Transfer Data

Storage

Object Detection

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

​

Reference

1. M. N. Al Islam, T. B. Hassan and S. K. Khan, "A CNN-based approach to classify cricket bowlers based on their bowling actions," 2019 IEEE International Conference on Signal Processing, Information, Communication & Systems (SPICSCON), Dhaka, Bangladesh, 2019, pp. 130-134, doi: 10.1109/SPICSCON48833.2019.9065090..

​

1. A. Z. M. E. Chowdhury, M. S. Rahim and M. A. U. Rahman, "Application of computer vision in Cricket: Foot overstep no-ball detection," 2016 3rd International Conference on Electrical Engineering and Information Communication Technology (ICEEICT), Dhaka, 2016, pp. 1-5, doi: 10.1109/CEEICT.2016.7873086.

​

1. P. Cika, M. Zukal, Z. Libis and M. K. Dutta, "Tracking and speed estimation of selected object in video sequence," 2013 36th International Conference on Telecommunications and Signal Processing (TSP), Rome, Italy, 2013, pp. 881-884, doi: 10.1109/TSP.2013.6614066.

​

IT18190762 | A.M.Karthik | PROJECT ID - 2021-172

​

​

A.Narthanan | IT18188196

BSc Hons information Technology specialization in Software Engineering

​

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

INTRODUCTION

• Identify the wicket taking percentage of the user's ball​
• Providing suggestion to increase the wicket taking percentage

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

RESEARCH PROBLEM

• There is no application to identify the wicket taking percentage of the user.​
• There is no application to provide suggestion to improve the wicket taking percentage.

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

RESEARCH QUESTIONS

• Where to get the dataset?​
• What are the features that influence the wicket taking percentage?​
• How to provide the wicket taking percentage to the user?​
• How to provide suggestions to increase the wicket taking percentage?

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

OBJECTIVES

Main Objective

• Providing the wicket taking percentage of the user's ball.​
• Providing suggestions to improve the wicket taking percentage.

• Create a custom dataset.​
• Provide the wicket taking percentage.
• Provide suggestions to increase the wicket taking percentage.

Sub Objective

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

COMPONENT DIAGRAM

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

TRAINING CUSTOM DATASET

Step 01: Get an image of a wicket taking ball when the ball hits the ground. ​

Source – wicket taking youtube videos with speed

​

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

Step 02: Extract batsman's x, y coordinate and ball's x,y coordinate from the image and speed of the ball​

Used Darknet YOLO object detection algorithm to extract the coordinates. ​

Speed of the video is available in the youtube video.

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

Dataset

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

Features

1. X coordinate of the batsman
2. Y coordinate of the batsman
3. X coordinate of the ball
4. Y coordinate of the ball
5. Speed of the ball

Labels

​

1. 0 - non-wicket taking
2. 1 - wicket taking

METHODOLOGY

What are x and y coordinates?​

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

0 - Coordinates of the ball​

1 – Coordinates of the batsman

​

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

Model Accuracy

TECHNOLOGIES

App Development

Manipulate data

Transfer Data

Storage

Object Detection

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

TECHNOLOGIES

App Development

Manipulate data

Transfer Data

Cricket Shot Detection

ML Library

Ball Detection

Storage

Test

1. S. Mane and S. Mangale, "Moving Object Detection and Tracking Using Convolutional Neural Networks," 2018 Second International Conference on Intelligent Computing and Control Systems (ICICCS), Madurai, India, 2018, pp. 1809-1813, doi: 10.1109/ICCONS.2018.8662921.​

​

2. Enrique J. Fernandez-Sanchez*, Javier Diaz and Eduardo Ros, "Background Subtraction Based on Color and Depth Using Active Sensors", Sensors, vol. 13, pp. 8895-8915, 2013.

​

REFERENCES

IT18188196 | A.Narthanan | PROJECT ID - 2021-172

​

​

BUSINESS MODEL

TRIAL

PREMIUM

FREE

$1.99/m

• 7 days of usage
• International cricket shots limited
• Wicket taking percentage

​

• Access to all international players
• Speed of the ball
• Assisting users

THANK YOU