0% found this document useful (0 votes)
117 views7 pages

Continuous Gesture Recognition For Flexible Human-Robot Interaction

This document summarizes a student project on continuous gesture recognition for human-robot interaction. It outlines the timeline of the project, which included literature review, developing algorithms for gesture recognition and classification using CNN and LSTM, testing and integrating the hardware and software, and future documentation. It also discusses progress made over vacations, including interfacing a camera with a Raspberry Pi and testing pre-trained models with poor initial accuracy, leading to collecting their own gesture dataset to train a CNN model.

Uploaded by

muhammad bilal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
117 views7 pages

Continuous Gesture Recognition For Flexible Human-Robot Interaction

This document summarizes a student project on continuous gesture recognition for human-robot interaction. It outlines the timeline of the project, which included literature review, developing algorithms for gesture recognition and classification using CNN and LSTM, testing and integrating the hardware and software, and future documentation. It also discusses progress made over vacations, including interfacing a camera with a Raspberry Pi and testing pre-trained models with poor initial accuracy, leading to collecting their own gesture dataset to train a CNN model.

Uploaded by

muhammad bilal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 7

Continuous gesture recognition for

flexible human-robot interaction

Presented By
Muhammad Bilal 16-ECT-50
Muhammad Yasin 16-ECT-78
Bilal Mustafa 16-ECT-08

Project Supervisor: Dr. Furqan Shaukat


Outline of Presentation
 Time Line
 Progress in Vacations

Department of Electronics Engineering


Time Line(1/3)
7th Semester 2019-20

Phase No Phase Estimated Time Date(2019-20) Status

1 Literature Review 6 Weeks 01-09 to 12-10 Completed

2 Algorithm for gesture 6 Weeks 13-10 to 24-11 Completed


Recognition (CNN)

3 Algorithm for gesture 4 Weeks 25-11 to 23-12 Completed


classification (LSTM)

Department of Electronics Engineering


Time Line(2/3)
8th Semester 2020
Phase No Phase Estimated Time Date(2020) Status

1 Testing ,Verification and 2 Weeks 01-02 to 15-02 Completed


modification of Algorithm

1 Purchase Hardware(Robot) 2 Weeks 16-02 to 29-02 Completed

2 Designing Robot 3 Weeks 01-03 to 21-03 Completed


Hardware
3 Integration of hardware 2 Weeks 22-03 to 04-04 In
and software Progress
4 Testing and Debugging 2 Weeks 05-04 ---- In
Progress
Department of Electronics Engineering
Time Line(3/3)
8th Semester 2020

Phase No Phase Estimated Time Date(2020) Status

5 Documentation 2 Weeks ------- ----

6 Documentation Review 1 Weeks ------ ----

7 Submission -----

Department of Electronics Engineering


Progress in Vacations
 We interface raspberry pi with camera , this camera is specially designed for
Rasberry Pi

 We test our pre trained model on raspberry pi , (We have not complete hardware
because, Robotic arm is in hostel because when vacations announced we did not
know exact duration at that time)

 Initially the accuracy is very poor, It was very rare when it detect accurate gesture

Department of Electronics Engineering


Progress in Vacations
 We make little changes code in, because it was designed to control keyboard, But we need to take
different signals for servo motors

 We write a code using open cv and some Image processing concepts to detect gestures, It detect
gesture on basis of defect points. In good lighting conditions it can static gestures .

 We also test some other models to detect gesture from github.

 Sir now we are decide to train CNN model on our own Dataset , we are making videos of only
five gestures .

Department of Electronics Engineering

You might also like