KARNATAKA STATE COUNCIL FOR SCIENCE AND TECHNOLOGY

Indian Institute of Science campus, Bengaluru

Telephone: 080 -23600978, 23341652 || Email: [email protected]
Website: www.kscst.org.in/spp.html or https://kscst.karnataka.gov.in/en

FORMAT FOR STUDENT PROJECT PROPOSAL FOR THE

48th SERIES OF STUDENT PROJECT PROGRAMME

(Handwritten proposals will not be accepted, please fill all the details in this MS word file, insert images / diagrams
wherever necessary. Convert to pdf file, get it approved from the project guide / head of the department and principal of
your institution. Keep ready the scanned pdf file of 1) Declaration and Endorsement 2) details of processing fees made
and fill-up the Google Form.
https://forms.gle/ks2WxWB4ei1hgv9D9

1. Name of the College: Jain College of Engineering and Technology,

Hubli

2. Project Title: AI Based Image Caption Generator In Regional Language

3. Branch: Computer Science And Engineering

4. Theme (as per KSCST poster): Data Science, Cyber Security,

Computing, Pattern Recognition and Image Processing, Signal
Processing.
5. Name(s) of project guide(s):
Name: Prof. Vishwanath Hiregoudar
Email id: [email protected]
Contact No.: 7411531675

6. Name of Team Members (Strictly not more than four students in a

batch):

Name: AMAAN FAROQUI

USN No.: 2JH21CS012
Email id: [email protected]
Mobile No: 8095445786

KSCST: Student Project Programme: 48th series: 2024-2025 1

 Name: AMIT BHANDARE
USN No.: 2JH21CS013
Email id: [email protected]
Mobile No.: 7760317732

Name: ASHISH KUNTHE

USN No.: 2JH21CS022
Email id: [email protected]
Mobile No.: 9480916537
7. Team Leader of the Project:
Name: ASHISH KUNTHE
USN No.: 2JH21CS022
Email id: [email protected]
Mobile No.: 9480916537
8. Processing Fee Details (Through Online Payment only):
(processing fee of Rs. 1180/-)
Transaction ID: T2411281448301348772855
UTR: 987573779791
Date: 28-11-2024
Time: 02:48 PM
Bank Name: State Bank of India

9. Date of commencement of the Project: 7-10-2024

10. Probable date of completion of the project: 5-3-2024

11. Scope / Objectives of the project:

Scope of the Project
1. Image Input Handling:
o Enable users to upload images directly or provide image URLs for caption
generation.
o Support various image formats (e.g., JPG, PNG, GIF).
2. Real-Time Caption Generation:
o Leverage AI/ML models (e.g., TensorFlow, PyTorch, or pre-trained models like
BLIP or CLIP) to process images and generate accurate captions.
o Ensure real-time response with minimal latency for better user experience.

KSCST: Student Project Programme: 48th series: 2024-2025 2

 3. Interactive User Interface:
o Develop a visually appealing and intuitive UI using Next.js and Tailwind CSS.
o Allow users to view generated captions immediately alongside the uploaded
image.
4. Scalability and Performance:
o Implement back-end optimization for handling large-scale image data.
o Use scalable cloud platforms or APIs for efficient AI model deployment.
5. Additional Features (Optional):
o Multi-language support for captions.
o Provide additional tags or descriptive keywords for SEO purposes.

Objectives of the Project

o Automate Image Captioning: Create an application that uses AI to generate

meaningful, contextually accurate captions for images.
o Real-Time Processing: Ensure the system processes images and generates
captions in real-time or near real-time.
o User Engagement: Build a seamless and engaging experience for users who need
quick captions for personal or professional use.
o Learning and Skill Development:

12. Methodology:

1. Project Planning and Requirement Gathering

 Define project goals, scope, and deliverables (e.g., real-time captioning, responsive UI).
 Identify required technologies and tools:
o Frontend: Next.js, React.js, Tailwind CSS.
o Backend: Next.js API routes or a separate back-end framework.
o AI Model: Pre-trained models like BLIP (Bootstrapped Language-Image
Pretraining) or CLIP (Contrastive Language-Image Pretraining).
 Select hosting and deployment platform (e.g., Vercel, AWS, or Google Cloud Platform).

2. Design and Architecture

 Frontend Design:
o Design a responsive UI using Figma or other design tools.
o Focus on user-friendly interactions for image uploads and displaying captions.
 Backend Architecture:
o Create API endpoints for processing image data and fetching captions.
o Integrate AI models for caption generation via libraries (e.g., PyTorch, TensorFlow,
or Hugging Face).

KSCST: Student Project Programme: 48th series: 2024-2025 3

 3. Data Processing and Model Integration
 Select a pre-trained AI model for image-to-text tasks.
o Example: Hugging Face’s BLIP or Image Captioning Transformer models.
 Integrate the model using a Python back-end (e.g., FastAPI) or JavaScript-based libraries.

4. Development Phases
 Frontend Development:
o Build the UI with Next.js and style it using Tailwind CSS.
o Implement file upload functionality for users to upload images.
o Display generated captions in real-time or after processing.
 Backend Development:
o Create APIs to handle image uploads and interact with the AI model.
o Process images using the AI model to generate captions.
o Send captions back to the frontend for display.

5. Testing and Debugging

 Unit Testing: Test individual components (e.g., image upload, caption rendering).
 Integration Testing: Ensure smooth interaction between frontend and backend.
 Performance Testing:
o Measure response times for caption generation.
o Optimize model inference for better real-time performance.
 Cross-Browser and Device Testing: Ensure compatibility across different browsers and
devices.

6. Deployment and Hosting

 Deploy the frontend and backend using platforms like Vercel or AWS Lambda.
 Use a CDN for faster image uploads and delivery.
 Monitor performance and uptime using tools like Google Cloud Monitoring or Postman.

7. Documentation
 Code structure and comments.
 User manual for the application.
 Deployment instructions for future scalability.

8. Feedback and Iteration

 Gather feedback from users and stakeholders.
 Incorporate improvements based on usability and performance.

9. Future Scope and Enhancements

 Add advanced features like multi-language support or image editing tools.
 Upgrade the AI model for more complex and descriptive captions.
 Integrate social media sharing for images and captions.

KSCST: Student Project Programme: 48th series: 2024-2025 4

13. Expected Outcome of the project:
A web application where users can:

 Upload images directly from their devices.

 Provide image URLs for caption generation.
 Automatic generation of descriptive and contextually accurate captions for
uploaded images.
 Real-time or near real-time caption generation with minimal processing delay.
 Display of generated captions alongside the uploaded image in a clear and user-
friendly interface.
 Support for various common image formats such as JPG, PNG, and GIF.
 Seamless interaction between the frontend and backend for image processing and
caption retrieval.
 Responsive design ensuring smooth usage across devices (desktop, tablet, and
mobile).

14. Is the project proposed relevant to the Industry / Society or

Institution?

Yes / No: No
If yes, please provide details of the industry / institution and contact
details:

15. Can the product or process developed in the project be taken up for
filing a Patent?
Yes / No: No

Prior Art search done?

Yes/No: No
Note: If your answer is “Yes”, you may contact Patent Information Centre of
KSCST. For more details, email: [email protected]

KSCST: Student Project Programme: 48th series: 2024-2025 5

16. Budget details (break-up details should be given):
Note: KSCST will provide nominal grant support for carrying out the project by students if
selected by the project selection committee.

Budget Amount

a) Materials / Consumables (Image 2500.00

Dataset)

b) Labor (Describe) 500.00

c) Travel (Testing Environment) 1000.00

e) Miscellaneous (API cost) 2000.00

Total 6000.00
17. Any other technical details (Please specify):

 Frontend Design:
o Design a responsive UI using Figma or other design tools.
o Focus on user-friendly interactions for image uploads and displaying captions.
 Backend Architecture:
o Create API endpoints for processing image data and fetching captions.
o Integrate AI models for caption generation via libraries (e.g., PyTorch, TensorFlow,
or Hugging Face).

18. SPP Coordinator (Identified by the college):

Name: Prof. Vishwanath Hiregoudar

Email id: [email protected]
Contact No.:7411531675

Guide: Prof. Vishwanath Hiregoudar HOD: Dr. Maheshkumar Patil

Email id: [email protected] Email id: [email protected]
Contact No.: 7411531675 Contact No.:7204113920

KSCST: Student Project Programme: 48th series: 2024-2025 6

 DECLARATION

We, the project team hereby declare that the details enclosed in the project proposal Title
of the Project: AI Based Image Caption Generator In Regional Language
Branch: Computer Science. College :Jain College of Engineering and Technology.. are
true and correct to the best of our knowledge and belief. We undertake to inform the
Karnataka State Council for Science and Technology (KSCST) of any changes to the
project title or team members' names immediately through our project guide or the SPP
Coordinator of our institution.
Additionally, we declare that the project work is original and not a result of copying or
purchasing. We are committed to completing the project independently, with support from
our faculty and project guide, while utilizing the facilities provided by the college. We will
not engage in plagiarism and pledge to be sincere and dedicated in executing and
completing the project as proposed.
We understand that if any of the above information is found to be false, untrue, or
misleading, we may be held liable. We authorize the sharing of the project information
contained in this proposal with KSCST, Bengaluru.
We acknowledge that the project team must exhibit and demonstrate the project,
participate in the mid-term evaluation of sanctioned projects, and engage with experts.
Additionally, we must exhibit the project at the Annual State-Level Poster Presentation and
Exhibition, if selected. Should our team fail to participate in the mid-term evaluation or the
Annual Exhibition (if selected), we understand that the supported project funds will be
returned to KSCST.
We also enclose the endorsement form for KSCST, Bengaluru.
Name of the students with USN No. Signature with date

1. AMAAN FAROOQUI 2JH21CS012

2. AMIT BHANDARE 2JH21CS013

3. ASHISH KUNTHE 2JH21CS022

Prof. Vishwanath HIREGOUDAR HOD: Dr. MAHESHKUMAR PATIL

Email id: [email protected] Email id: [email protected]
Contact No.: 7411531675 Contact No.: 7204113920

KSCST: Student Project Programme: 48th series: 2024-2025 7

 ENDORSEMENT

This is to certify that 1) Mr. AMAAN FAROQUI, 2) Mr. AMIT BHANDARE

3) Mr. ASHISH KUNTHE ,are bonafide student(s) of Department of Computer Science And
Engineering , in the degree program of our institution. If the project proposal submitted by
these students under the 48th series of Student Project Programme is selected by KSCST,
we will provide the requisite laboratory / Computer / infrastructure support in our college /
Institution. Further we also take necessary steps to see that the project team will exhibit /
demonstrate their project in the mid-term evaluation of project and in the Annual State-
Level Poster Presentation and Exhibition (if selected). If the student team fails to send the
completed project report or fails to attend the evaluation in mid-term evaluation of
sanctioned projects or fails to attend the Annual State-Level Poster Presentation and
Exhibition (if selected), the supported project amount will be returned to KSCST.

Prof. VISHWANATH HIREGOUDAR Dr. MAHESHKUMAR PATIL Dr. PRASHANTH BANAKAR

Emailid: Emailid: Emailid:

[email protected] [email protected]
[email protected]

Contact No.: 7411531675 Contact No.: 7204113920 Contact No.:9900114491

KSCST: Student Project Programme: 48th series: 2024-2025 8

 DETAILS OF PROCESSING FEES MADE THROUGH
NEFT / UPI PAYMENT

1. TITLE OF THE PROJECT : AI image Caption generator in native

language

2. NAME OF THE TEAM LEADER : ASHISH KUNTHE

3. EMAIL ID : [email protected]

4. CONTACT MOBILE NO. : 9480916537

PAYMENT MADE DETAILS

5. BANK REF. NO. / UTR NO. / : 8095445786

UPI No. (12 digits)

6. TRANSACTION ID : T2411281448301348772855

7. NAME OF THE SENDER / : AMAAN HUSSAIN FAROQUI

ACCOUNT HOLDER and
8095445786
CONTACT NUMBER

8. NAME OF THE BANK : STATE BANK OF INDIA

9. PROCESSING FEES : Rs. 1,180/- (Inclusive of 18% GST)

10. DATE OF PAYMENT MADE : 28/11/2024

11. TIME : 2:48 PM

12. MODE OF PAYMENT MADE : UPI

(NEFT / UPI, PLEASE SPECIFY)

(ASHISH KUNTHE) (Prof. Vishwanath Hiregoudar)

(Signature of (Name & Signature of
the team leader) Project Guide with Seal)

KSCST: Student Project Programme: 48th series: 2024-2025 9

 HOD: Dr. MAHESHKUMAR PATIL

Email Id: [email protected]

Contact No. : 7204113920

KSCST: Student Project Programme: 48th series: 2024-2025 10