SIGN LANGUAGE

DETECTION

Presented by
Meghana Kanigolla Mentor
Shravani Khodave Prof.AfrinSheikh
Siddhi Koli
Vaishnavi Kusekar
Introduction
• Sign language is a key mode of communication for the deaf
community.
• ML can help translate hand signs into text to improve
accessibility.
• Our project detects a hand sign from a frame and displays
the corresponding text.
 Literature review
SR. NO NAME OF TITLE OF CLAIM BY Future
THE PAPER AUTHOR Scope
AUTHOR
1 Hamza Mnassri; Sign Language This paper explores To improve the
Riadh Bchir; Detection Based on the intersection of accuracy with larger
Mohamed Amine Artificial Intelligence Artificial Intelligence datasets and better
Zayane; Taoufik Ladh from Images (AI) and sign feature extraction
ari language
communication,
aiming to enhance
accessibility
2 Jeet Debnath; RealTime Gesture This research We want to further
Praveen Joe I R Based Sign Language presents a robust and enhance, improve,
Recogniti on System real-time Gesture- and real-world
Based Sign Language implementation
Detection System
that leverages
computer vision and
3 Kajal Dakhare; A Survey on deepLSTM
The learning
model To make the
Vidhi Wankhede; Recogniti on and techniques.
applied towards recognition more
Prateek Verma Translati on System understanding sign efficient in future
of Real Time Sign language gestures by using larger
Language in real time datasets
4 Gladys Jessica Ruslim; Sign Language This study intends to To ensure clear hand
Nicholas Matthew Detection to Enhance improve these sign images in the
Salim; Ivan Sebastian Online people's online dataset, and develop
Edbert; Derwin Communication for communication the application of sign
Suhartono Deaf and Mute experiences language
Individuals through
Deep Learning Models
Problem Statement

• Many people cannot understand sign language, creating a

communication gap.
• Existing translation methods are not widely available.
• Our solution provides real-time, automated sign recognition .
Project Concept

• A single frame is analyzed to detect a hand sign.

• The system recognizes the sign and converts it into text.
• The detected sign and corresponding text are displayed in
real time.
Methodology

• Frame Capture:
- Extract a single frame from video.
• Hand Detection:
- Use of OpenCV to locate the hand.
• Feature Extraction:
- Identify hand landmarks and gestures.
• Sign Recognition:
- Tensorflow object detection to detect the sign.
• Text Output:
- Display corresponding text below the detected sign.
System Architecture

• Input:
-Single video frame with hand sign.

• Processing:
- Hand detection using OpenCV
- Feature extraction using ML model
- Sign classification by Tensorflow object detection api.

• Output:
- Display detected sign with text.
Workflow Diagram

• [Insert diagram showing step-by-step process: Frame → Hand

Detection → Sign Recognition → Text Output]
Results

• The model correctly detects and labels hand signs from video
frames.
• Text output is displayed below the detected sign.
• Accuracy improves with larger datasets and better feature
extraction.
• Real-time sign detection and translation can improve
accessibility.
Applications

• Real-time ML sign language translators.

• Assistive technology for the hearing-impaired.
• Education tools for sign language learning.
Future Enhancements

• Improve model accuracy with larger datasets.

• Integrate real-time translation with voice output.
• Extend support for multiple sign languages.
REFERENCES

[1] Hamza Mnassri; Riadh Bchir; Mohamed Amine Zayane; Taoufik Ladh
ari,Sign Language Detection Based on Artificial Intelligence from
Images,2024 IEEE International Conference on Artificial Intelligence &
Green Energy (ICAIGE)

[2] Jeet Debnath; Praveen Joe I R,RealTime Gesture Based Sign Language
Recogniti on System,2024 International Conference on Advances in Data
Engineering and Intelligent Computing Systems (ADICS)

[3] Kajal Dakhare;Vidhi Wankhede;Prateek Verma A Survey on

Recognition and Translati on System of Real-Time Sign Language,2024
2nd DMIHER International Conference on Artificial Intelligence in
Healthcare, Education and Industry (IDICAIEI)

[4] Gladys Jessica Ruslim; Nicholas Matthew Salim; Ivan Sebastian

Edbert; Derwin Suhartono,Sign Language Detection to Enhance Online
Communication for Deaf and Mute Individuals through Deep Learning
Models,2024 International Conference on Computer Engineering,
Network, and Intelligent Multimedia (CENIM) jal Dakhare; Vidhi
Wankhede; Prateek VermaKajal Dakhare; Vidhi Wankhede;
PrateeKajal Dakhare; Vidhi Wankhede; Prateek Veajal Dakhare;
Vidhi Wankhede; Prateek Verma
Some sign languages