Project Request: Tuberculosis Detection from X-ray Images

Problem Statement

Tuberculosis (TB) is a highly infectious disease that primarily affects the lungs. Early detection and diagnosis of TB are crucial for effective treatment and disease control. X-ray imaging is one of the commonly used methods to diagnose TB, where radiologists analyze chest X-ray images to identify the presence of TB-related abnormalities.

The goal of this project is to develop an automated system for tuberculosis detection using chest X-ray images. The system will utilize machine learning and image processing techniques to analyze X-ray images and classify them as TB-positive or TB-negative.

Project Scope

The project will involve the following key tasks:

1. Data Collection: Gather a large dataset of chest X-ray images, consisting of both TB-positive and TB-negative cases, with proper annotations indicating the presence or absence of tuberculosis.

2. Data Preprocessing: Perform necessary preprocessing steps on the collected X-ray images, such as resizing, normalization, and augmentation, to ensure uniformity and improve model performance.

3. Model Development: Build a deep learning model, such as a convolutional neural network (CNN), to learn from the X-ray images and classify them as TB-positive or TB-negative. The model should be trained using the annotated dataset and optimized for high accuracy and sensitivity.

4. Model Evaluation: Evaluate the trained model using appropriate evaluation metrics, such as accuracy, precision, recall, and F1-score, to assess its performance in tuberculosis detection. Cross-validation or other techniques should be employed to ensure robustness and generalization of the model.

Skills Required

The following skills are required for the successful completion of this project:

• Strong knowledge of machine learning and deep learning techniques, particularly in the field of computer vision.
• Experience in image processing and analysis, specifically with medical images.
• Proficiency in Python programming and popular deep learning libraries, such as TensorFlow or PyTorch.
• Familiarity with data preprocessing techniques for image data and handling large datasets.

Resources

The project will require a dedicated team consisting of the following roles:

• Machine Learning Engineer: Responsible for data preprocessing, model development, and evaluation.
• Data Annotation Specialist: Required during the data collection phase to annotate the X-ray images with tuberculosis labels.

Conclusion

The successful completion of this project will contribute to the early detection and diagnosis of tuberculosis, aiding healthcare professionals in providing timely treatment and reducing the spread of the disease. The automated system will provide a valuable tool

Project Request

Define You

• GSSOC Participant
• Contributor

Project Name

Description

This model will detect colors in an image

Please assign it to me @adithya-s-k

Project Request

Define You

• [ .] GSSOC Participant
• [ .] Contributor

Project Name

Description

Perform complete EDA to get insights
Use Sampling Techniques
We can implement few deep learning models in order to increase the efficiency of the project to get better outcomes.

Scope

Use Case : real time sentiment analysis

Timeline

[The project's estimated start and end dates, milestones, and deadlines for deliverables]
3 weeks from the date of approval.

Video Links or Support Links

[Links that can support the project in anyway]

Project Request

Define You

• GSSOC Participant
• Contributor

Project Name

The Sign Language Recognizer

Description

Sign language recognition using deep learning and machine learning is a technology that aims to bridge the communication gap between individuals who are deaf or hard of hearing and those who use spoken languages. It involves the development of algorithms and models that can interpret and understand sign language gestures, allowing for real-time translation and communication.
The goals of a sign language recognition project using deep learning and machine learning can vary depending on the specific context and objectives. However, here are some common goals that such a project may aim to achieve:

Accurate Gesture Recognition: The primary goal is to develop a system that can accurately recognize and interpret sign language gestures in real time. The system should be able to identify a wide range of sign language gestures, including hand shapes, movements, and facial expressions, with a high level of accuracy.

Real-Time Performance: Another important goal is to ensure that the sign language recognition system operates in real time, providing instantaneous feedback and translation. This enables smooth and natural communication between individuals using sign language and those who use spoken languages.

Robustness and Adaptability: The system should be robust and adaptable to different users, lighting conditions, camera angles, and environmental factors. It should be able to handle variations in signing styles, individual differences in gestures, and accommodate different sign language dialects or variations.

Scalability and Accessibility: The project may aim to develop a scalable and accessible sign language recognition system that can be easily deployed and used in various settings. This includes integration with different devices and platforms, such as smartphones, tablets, or wearable devices, to ensure widespread accessibility.

User-Friendly Interface: The project may focus on designing a user-friendly interface that facilitates intuitive interaction and communication for individuals with hearing impairments. The interface should be easy to navigate, provide clear visual feedback, and support additional features such as text-to-speech or sign-to-speech translation.

Dataset Creation and Expansion: Building a comprehensive and diverse dataset of sign language gestures is often a goal in these projects. This involves collecting a large and representative set of sign language samples to train and evaluate the deep learning models effectively. The project may also contribute to expanding existing sign language datasets or developing new ones to support further research in the field.

Integration with Assistive Technology: The project may aim to integrate the sign language recognition system with other assistive technologies, such as augmented reality glasses, haptic feedback devices, or voice recognition systems. This integration can enhance the overall user experience and provide more comprehensive support for individuals with hearing impairments.

Continuous Improvement: Continuous improvement and refinement of the sign language recognition system is an ongoing goal. This includes refining the algorithms, optimizing model performance, addressing any limitations or challenges, and incorporating user feedback to enhance the system's accuracy, usability, and overall effectiveness.

Overall, the goals of a sign language recognition project using deep learning and machine learning revolve around developing an accurate, real-time, and user-friendly system that improves communication accessibility for individuals with hearing impairments, fostering inclusivity and empowering them in various aspects of life.

Scope

The scope of sign language recognition using deep learning and machine learning is vast and holds significant potential in various domains. Here are some key areas where this technology can have a profound impact:

Communication Accessibility: Sign language recognition systems can enable seamless communication between individuals who are deaf or hard of hearing and those who use spoken languages. This technology has the potential to break down barriers and promote inclusivity in educational, professional, and social settings.

Education and Learning: Sign language recognition can enhance the accessibility of education for individuals with hearing impairments. It can be integrated into e-learning platforms, allowing students to access sign language interpretation during online courses, lectures, or video tutorials. This empowers students to learn at their own pace and effectively engage with educational materials.

Assistive Technology: Deep learning-based sign language recognition can be integrated into assistive devices, such as wearable devices, smartphones, or tablets, to facilitate real-time communication for individuals with hearing impairments. This technology can help them communicate with hearing individuals, access public services, and navigate everyday situations more independently.

Human-Computer Interaction: Sign language recognition can enable more intuitive and natural interaction between humans and computers. It can be used to develop sign language interfaces that allow individuals with hearing impairments to control various devices, access information, and interact with digital applications using sign language gestures.

Sign Language Translation: Deep learning-based sign language recognition can be coupled with machine translation techniques to enable real-time translation between sign language and spoken or written languages. This technology can facilitate communication between individuals who use different languages and bridge the language gap for deaf or hard of hearing individuals in multicultural or international settings.

Accessibility in Media and Entertainment: Sign language recognition can be applied to improve accessibility in media and entertainment. It can be used to automatically generate sign language interpretations or subtitles for video content, making movies, TV shows, and online videos more inclusive and enjoyable for individuals with hearing impairments.

Research and Development: Sign language recognition using deep learning and machine learning presents ample opportunities for research and development. Researchers can explore new algorithms, architectures, and datasets to improve the accuracy, robustness, and efficiency of sign language recognition systems.

Timeline

22nd May to 10Aug

Video Links or Support Links

[Links that can support the project in anyway]

Project Request

Book Recommendation Model - A recommendation model that would help in recommending books.
It has 2 ways -
Collaborative filtering
Content based filtering

https://github.com/Shreyg-27

Define You

• GSSOC Participant
• Contributor

Project Name

Book Recommendation System

Description

A book recommendation system that can either be collaborative or content based. So a person should be able to recommend a book based on either if the 2 features mentioned.

Scope

To recommend a book based on either collaborative/ content based

Timeline

Date of assign - August 1 2023

Project Request

A machine learning model that predicts the fuel consumption of a car.

GitHub: https://github.com/Praneesh-Sharma

Define You

• [✔️ ] GSSOC Participant
• [ ✔️] Contributor

Project Name

Fuel Consumption Prediction

Description

Gives the user the amount of fuel (in gallons) consumed by a car. The parameters to be entered by the user will be Cylinders, Displacement, Horsepower, Weight, Acceleration and Model Year. The output will be the miles covered per gallon.

Scope

This model will help the user decide on which car is more economical and which cars to consider while deciding to buy a car.

Timeline

Start Date: Date of Project Assignment
End Date: 1 week from Start Date

Crop Production Prediction using ML techniques

Predicting the crop production for various crops using ML regression techniques.
Dataset: https://www.kaggle.com/datasets/abhinand05/crop-production-in-india

https://github.com/asthabarwal

I am a

• [*] GSSOC Participant

@adithya-s-k I would love to work on this issue

Project Name

Crop Yield Prediction

Description

Gives user info about produce of crops based on their given information

Scope

Aims to predict the crop produce based on the season, area, and district etc

Timeline

Start date: Project assignment
End date: 6 June

###Extra info

Please assign this issue to me under GSSOC'23

Project Request

Github: https://github.com/Durga-Sowjanya-Sanku

Define You

• [✔️] GSSOC Participant
• [ ✔️] Contributor

Project Name

Fake News Detection

Description

The dataset Consists of the text, title, subject and date and we need to predict if that is a fake news or not.

Scope

This model helps user to understand which is fake news and which is not.
Also helps the learner to deal with such datasets

Timeline

Start Date : Date of Assigning
End Date : Start Date + 9 Days

Project Request

https://github.com/Danie-O

Define You

• GSSOC Participant
• Contributor

Gender Prediction

Description

Gender prediction using image classification

Scope

A gender prediction system that predicts the gender of an individual by identifying features from their images. The system can be deployed as a Streamlit app through a prediction function.

Timeline

End date: 20th June, 2023.

Project Request

Define You

• GSSOC Participant
• Contributor

Real-time Sign Language Detection System

Description

This project will help people who cannot speak, to effectively communicate their thoughts to those who cannot understand sign language, thus inducing bilateral talks. This will also help people get familiar with sign language because interacting with new people teaches us new skills.

Scope

One of the major constraints of this project is that if the signs are displayed too fast then the system may lag but further optimizations and improvements will remove this problem too.

Timeline

The project will take almost a week to get accomplished.

Video Links or Support Links

Links that can support the project in anyway

Project Request

The goal of this project is to develop an AI-powered system that leverages natural language processing (NLP) and computer vision techniques to generate and manipulate text and images. The system will have multiple functionalities, including text completion and editing, text embeddings for search and classification, code generation and editing, and image generation and editing.

The project will involve building and training deep learning models for each specific task, integrating them into a cohesive system, and developing a user-friendly interface to interact with the system's functionalities. The system will be designed to handle various text and image data types and provide accurate and efficient results.

Define You

• ☑ GSSOC Participant
• ☑ Contributor

AI-Driven Text and Image Generation

Description

The project will involve building and training deep learning models for each specific task, integrating them into a cohesive system, and developing a user-friendly interface to interact with the system's functionalities. The system will be designed to handle various text and image data types and provide accurate and efficient results

Goals
Implementing text completion and editing functionalities.
Implementing text embeddings for search and classification.

Outcomes
Enable users to automatically generate and modify code snippets, facilitating software development and code refactoring tasks.
Creating a cohesive system where NLP and computer vision functionalities seamlessly work together, providing a comprehensive toolkit for text and image processing.

Scope

• Researching and implementing state-of-the-art NLP models for text completion, editing, and embeddings.
• Researching and implementing computer vision models for image generation and editing.
• Integrating NLP and computer vision models into a unified system.
• Developing a user-friendly interface for interacting with the system's functionalities.
• Testing and evaluating the system's performance and accuracy.
• Documenting the project, including a user guide and technical documentation.

Tech Stack - Python, TensorFlow,Natural Language Toolkit, spaCy, or Hugging Face's Transformers,OpenCV,Markdown,PyTest,Django

Timeline

start date - the day project is assigned
end date - 10th august.

Support Links

https://www.tensorflow.org/guide
https://spacy.io/usage
https://huggingface.co/docs/transformers/index
https://docs.opencv.org/
https://docs.pytest.org/en/latest/
https://docs.djangoproject.com/en/4.2/

Project Request

The project aims to develop a Stock Market Trading Agent using Deep Reinforcement Learning.

Define You

• GSSOC Participant
• Contributor

Stock Market Trading Agent using Deep Reinforcement Learning

Description

The project involves developing an intelligent trading agent that utilizes Deep Reinforcement Learning techniques to make trading decisions in the stock market. The agent will learn from historical stock price data and use a reinforcement learning algorithm to optimize its trading strategy. The goal is to create a robust and profitable trading agent that can adapt to changing market conditions and make informed trading decisions.

Scope

Data preprocessing: The historical stock price data will be preprocessed to extract relevant features for training the trading agent.

Reinforcement Learning model: A Deep Reinforcement Learning model will be implemented to train the trading agent. The model will learn to maximize cumulative rewards by making buy/sell decisions based on the input features.

Trading strategy optimization: The trading agent will continuously optimize its trading strategy by adjusting its actions based on feedback from the market. This will involve exploring different trading policies and evaluating their performance.

Evaluation and analysis: The performance of the trading agent will be evaluated using various metrics, such as profitability, risk-adjusted returns, and comparison with benchmark strategies. The project will also include an analysis of the agent's behavior and decision-making process.

Timeline

End Date: as soon as possible(after assigned it to me)

Project Request

"SentinelAI: Harnessing Satellite Vision for Rapid Disaster Response and Life-saving Insights"

https://github.com/prithubanik

Define You

• GSSOC Participant
• [ Prithu Banik ] Contributor

Project Name

SentinelAI

Description

SentinelAI is an innovative project that combines the power of Convolutional Neural Networks (CNN), Generative Adversarial Networks (GAN), and Support Vector Machines (SVM) to predict and prioritize the most heavily impacted areas during disasters. By leveraging advanced machine learning techniques, SentinelAI aims to optimize rescue efforts and save lives by accurately identifying areas requiring urgent attention.

Scope

[Objectives: The project aims to develop and implement a CNN-based model to analyze satellite imagery and predict heavily impacted areas during natural disasters, facilitating efficient rescue operations and resource allocation.

Deliverables: The project will deliver a trained CNN model capable of identifying and classifying areas with severe damage or potential hazards, along with an integrated GAN and SVM model for data augmentation and prioritization. A user-friendly interface or API will also be provided for real-time input and prediction.

Constraints: The project's constraints include limited availability of labeled satellite imagery for training, potential variations in disaster scenarios, and computational resource limitations for model training and inference. Time constraints and the need for continuous model updates to adapt to evolving disaster situations are also considered.]

Timeline

[Start date : the assigned date
End date: August 10 2023]

Project Request

Predict a customer's credit loan status given basic information such as salary, occupation, gender, credit score, and so on.

https://github.com/Danie-O)

Define You

• GSSOC Participant
• Contributor

Project Name

Loan Credit Status Prediction

Description

Loan Credit Status Prediction using ML techniques such as Linear, Ridge and Lasso regression as well as classification with logistic regression.

Scope

A Streamlit application and a simple Flask app through which users can enter details and have their credit loan status prediction displayed.

Timeline

End date: 30th July, 2023.

Project Request

Develop an object detection model using YOLO. An image will be the input to the model and the model will broadly classify the image into the classes it has been trained on.

https://github.com/kimix7

Define You

• GSSOC Participant
• Contributor

Object detection using YOLO

Description

The object detection project using YOLO (You Only Look Once) focuses on utilizing deep learning techniques to detect and track various objects in images or real-time video streams. By training a YOLO model with a dataset that includes labelled images, the model learns to recognize and localize objects within new images or video frames. This project provides accurate and efficient detection of objects, including people.

Scope

Objectives:

• Implement a robust and efficient object detection system using YOLO.
• Enable accurate detection and localization of various objects, including specific target objects.
• Learning how YOLO works.

Deliverables

• A trained YOLO model capable of detecting and localizing objects accurately.
• Well-documented guidelines, including dataset preparation, training, inference, and any specific requirements.
• Efficient and real-time object detection on images or video frames.

Constraints

• Availability and quality of labelled training data can impact the model's performance.
• Achieving high accuracy requires sufficient training iterations and careful fine-tuning of the model's parameters. It may involve iterative experimentation and adjustment to optimize detection performance.

Timeline

Start Date: when assigned
End Date: 10 August

co2 emission prediction model

It's a ML model which predicts the CO2 emission of a vehicle, based on the parameters such as number of cylinders, size of engine etc.

github.com/srish-ty

Define You

• GSSOC Participant
• Contributor

I'm a gssoc'23 contributor and
I request that please accept this proposal

Project Request

Fruit Image Classifier

This ML model would classify a fruit image to the respective fruit class

https://github.com/lcs2022026

• [ yes] GSSOC'23 Participant
• [yes ] Contributor

Problem Statement : To recognize classroom activities . I propose a Machine Learning model that can recognize activities inside a classroom from the Dataset that is collected online

Project Request

To use SVM to build and train a model using cell records, and classify cells are benign (mild state) or malignant (evil state).

|
| Github |NisargPipaliya |
| Email | [email protected] |
| Label | Project Request |

https://github.com/NisargPipaliya

Define You

• GSSOC Participant
• Contributor

Project Name

Cancer Cell Detection System

Description

[Description of the project, its goals, and expected outcomes]

Scope

[The project's boundaries, including its objectives, deliverables, and constraints]

Timeline

[The project's estimated start and end dates, milestones, and deadlines for deliverables]

Video Links or Support Links

[Links that can support the project in anyway]

Project Request

Define You

• GSSOC Participant
• Contributor

Project Name

Description

[SRGAN is a generative adversarial network for single image super-resolution. It uses a perceptual loss function which consists of an adversarial loss and a content loss. The adversarial loss pushes the solution to the natural image manifold using a discriminator network that is trained to differentiate between the super-resolved images and original photo-realistic images. Generating Images with better higher resolutions ]

Scope

[The project aims to improve the resolution of images with the help of a pytorch-based implementation.]

Timeline

[starting date: assignment date
duration: 5-10 days ]

Video Links or Support Links

[SRGAN]

Project Request

Image Captioning with Deep Learning

This project aims to develop a model that automatically generates descriptive captions for images.

https://github.com/aman-kumar29

Define You

• GSSOC Participant
• Contributor

Image Captioning with Deep Learning

Description

The project involves training a deep-learning model on a dataset of images paired with corresponding captions. The model will learn to understand the visual content of images and generate meaningful and contextually relevant captions. The project will also include evaluating the model's captioning performance and the development of a user-friendly interface for image caption generation.

[Description of the project, its goals, and expected outcomes]

Objectives

1. Real-Time Caption Generation: Develop a user interface where users can upload images, and the model generates captions in real time.
2. Content Generation: Image captioning models can be extended to generate creative and contextual captions for various applications, such as social media posts or advertising.

Deliverables

1. Will give a trained image captioning model which should be able to take an input image and generate a relevant and contextually appropriate caption that accurately describes the visual content.
2. A user-friendly interface to interact with the model
3. comprehensive documentation will be provided, including technical documentation and user guides which will clearly describe how to use the interface and how to generate the caption.

Timeline

Start Date: when assigned
End Date: 20 July

Support Links

1. http://cocodataset.org/ could use this to get a dataset for the same.
2. could go through some research papers for this

Project Request

An image classification model to classify PPI documents (Aadhaar Card, Pan Card, Driving License, Passport and Voter ID) using transfer learning.

https://github.com/nk-droid

Define You

• GSSOC Participant
• Contributor

PII Document Classification using transfer learning

Description

An image classification model to classify Personal Identifiable Information Documents using transfer learning. The model will classify if a document is Aadhaar Card, Pan Card, Driving License, Passport or Voter ID.

[Description of the project, its goals, and expected outcomes]

Scope

Objectives:

• Implement an efficient and accurate model using transfer learning and CNN.
• Learning the basics of image classification.

Deliverables

• A trained CNN model to identify PII documents.
• Properly documented notebook.

Constraints

• The performance of any model depends on the number of iterations and the amount of labelled data available. It may involve some iterative experiments to get desired results.

Timeline

Start Date: Date of assigning
End Date: 3 week after getting assigned

Project Request

Define You

• GSSOC Participant
• Contributor

Project Name

Solar Energy Output Prediction using ML

Description

The aim of the project is to predict solar energy output using ML using features related to location and weather conditions using simple linear regression.

Scope

Data exploration, feature engineering, training, and evaluation.

Timeline

Start date: 22nd May, 2023
Tentative end Date: 22nd June, 2023

Video Links or Support Links

Dataset: https://www.kaggle.com/datasets/saurabhshahane/northern-hemisphere-horizontal-photovoltaic

name: House price prediction
about: predict price of a house
title: "[PROJECT PROPOSAL]"

Project Request

Predice house price from given data

https://github.com/avijit1999

Define You

• GSSOC Participant
• Contributor

Project Name

House Prices prediction

Description

Here I want to develop a model to predict price of house using some data.

Scope

It is limited only data. It can mordify to webapp where user can enter the informations using keyboard.

Timeline

End Date: as soon as possible(after assigned it to me)

Image classification using Tensorflow

Define You

• Contributor

Image Classification using Tensorflow

This project classify the image

This project is the proof of work project. In this We can classify the image using tensorflow lbirary

Project Request

Develop an Collaborative Book Recommendation System that provide user with personalized book recommendation system. Users can input the name of a book, the author's name, and other relevant credentials. Based on this input, the system will generate a list of the top five most similar books according to the user's preferences, ratings, and popularity.

Define You

• GSSOC Participant
• Contributor

ReadMate - A Collaborative IntelliRec System

Description

A Collaborative Book Recommendation System empowers users to discover books based on their preferences and interests. This intelligent system utilizes advanced algorithms to provide personalized recommendations by analyzing user input. If a user enjoys reading books and desires to explore similar titles, this system proves to be invaluable. By inputting information about their preferred book and author, users can expect a curated selection of top-rated and popular books that align with their interests. This intelligent model caters to the user's specific choices and offers a seamless experience in finding new books to enjoy.

Scope

Objectives:

• Develop a Collaborative Book Recommendation System that leverages user preferences and behaviors to generate personalized
  book recommendations.
• Implement machine learning algorithms for analyzing user interactions, ratings, and reviews to identify patterns and similarities among users.
• Enhance the recommendation system by considering factors such as user ratings, book popularity, and author preferences to improve recommendation accuracy.
• Design and implement a user-friendly interface that allows users to input their book preferences and retrieve personalized recommendations easily.
• And the most important objective behind this project is to understand how a recommendation system can truly be helpful for users. Through this project, I aim to explore the effectiveness and benefits of a collaborative book recommendation system in providing users with personalized and relevant book recommendations based on their preferences and interests.

Constraints:

• The success of the recommendation system relies on the availability of a diverse and comprehensive dataset of books, user interactions, and ratings.

Timeline

Start Date: From Assign Date
End Date: After 2 weeks from Start Date

Project Request

Define You

• [✔️ ] GSSOC Participant
• [✔️] Contributor

Game of thrones personality matcher

Description

This is a project based on data science and machine learning which used TSNE algorithm and the main aim of this project is to match the personality of the game of thrones characters and this app is hosted on streamlit framework in pycharm IDE|

Timeline

In 3 days

Project Request

MOM creator extension for GMEET

www.github.com/samarthjain422005

Define You

GSSOC Participant
Contributor

MOM generator

Description

This project involves recording transcripts from google meets and then generate the MOM of the meeting , in order to make it easier for the users to comprehend the outcomes of the meeting. It uses NLP and other machine learning algorithms to generate the MOMs of the meetings.
If time permits i would like to extend the same feature as a discord bot for discord meetings , MS teams and Zoom meeting applications.

Scope

Objectives
-[ ] Implement a robust and efficient MOM generator.
-[ ] Convert it into a google extension.
-[ ] Enable high accuracy model using comparitive analysis of all machine learning models.

Deliverables

-[ ]A trained model capable of generating MOMs accurately.
-[ ]Well-documented guidelines, including dataset preparation, training, inference,exploratory data analysis ,etc.
-[ ]Models predicting using different algorithms of NLP , HMMs ,etc.
-[ ]An accurately google extension for deployment of the model.

Timeline

Start Date: On assigning
End Date: 30 July

Project Request

This project involves using deep learning techniques to develop an algorithm to identify the plant is infected by the disease or not

https://github.com/Shubhamkumar-op

Define You
I am
[ * ] GSSOC Participant

Plant Disease detection using CNN

Description

• I will use image classification using CNN using which a farmer can take a picture it will tell you if the plant has a disease or not
• In this I will use CNN model to classify image and detect disease

• The CNN model will be designed and trained using the collected dataset.
• Once the CNN model is trained, it will be tested using a separate set of images to evaluate its accuracy and performance.

Scope

• Implementing a CNN for potato disease detection aims to improve the accuracy, efficiency, and effectiveness of disease identification and management in potato farming, leading to higher crop productivity and reduced economic losses.
• After building the model woe can deploy a Mobile app so that it can be easily accessible to farmers

Timeline

Start time: Time of Assignment
End time: July 15th.

Project Request

Develop an AI system that can generate accurate and meaningful captions for images. The system will take an image as input and use computer vision techniques to understand the content of the image. It will then generate a descriptive caption that accurately represents the objects, actions, and context depicted in the image.

https://github.com/sangu-firedev

Define You

• GSSOC Participant
• Contributor

Project Name

AI image captioner

Description

An AI system which takes images as inputs and produces suitable captions. The model is trained on the dataset of images with human written captions. The project uses CNN models like ResNet or VGG to extract useful feautures from images, the extracted features is used as input for RNN models like LSTM or GRU which outputs sequences of words forming a caption. The generated output is evaluated using evaluation metrics like METEOR, BLEU or CIDEr measuring the quality and accuracy of captions, which helps in fine tuning the model. The model is deployed on streamlit with UI, where users can interact with the model and produce captions for their images.

Scope

objectives

• Develope a AI system which generates decriptive captions for images.
• Deploy the Model to upload images and receive captions instantly.

delivarables

• AI model to generate captions for images.
• source code for the project with documentation of implementation.
• application where users can interact with the model.

contraints

• The success of the projects depends upon the availability and quality of dataset.

Timeline

start date - the day project is assigned
end date - 10th august.

Project Request

Project Name

Description

[Description of the project, its goals, and expected outcomes]

Scope

[The project's boundaries, including its objectives, deliverables, and constraints]

Timeline

[As i have an experience of this project so it will take almost 4 to 7 days as i will amke some changes and will make it more attractive]

Video Links or Support Links

[Links that can support the project in anyway]

Sample

Other

I have added a sample of it I will improve it and will make it more attractive and datasets will be taken by using Kaggle. GSSOC'23 CONTRIBUTOR. @adithya-s-k @CoginitiveLab-tech

Project Request

Video Captioning with Deep Learning
The Video Captioning with Deep Learning project focuses on developing a model that automatically generates descriptive captions for videos.

https://github.com/aman-kumar29

Define You

• GSSOC Participant
• Contributor

Video Captioning with Deep Learning

Description

The project involves training a deep learning model on a labeled dataset of videos paired with corresponding captions. The model will learn to understand the visual content and temporal dynamics of videos and generate meaningful captions that describe the video content accurately. The project will also include the development of a user interface for real-time video caption generation and evaluation of the model's performance.

Scope

Objectives

1. Real-Time Caption Generation: Develop a user interface where users can upload videos, and the model generates captions in real time, providing a time-aligned description of the video content.
2. Content Discovery and Recommendation: Video captioning models can be integrated into video recommendation systems, enhancing personalized video recommendations based on user preferences and interests.

Deliverables

1. Will give a trained image captioning model which should be able to take an input video and generate a relevant and contextually appropriate caption that accurately describes the visual content.
2. A user-friendly interface to interact with the model
3. comprehensive documentation will be provided, including technical documentation and user guides which will clearly describe how to use the interface and how to generate the caption.

Timeline

Start Date: when assigned
End Date: 10 August

Video Links or Support Links

1. https://www.microsoft.com/en-us/research/project/msr-vdc-iccv-2013-video-to-text-challenge/ for the dataset. There is also ActivityNet Captions dataset.
2. Also some research papers would be helpful in choosing and changing the architecture of the model

Project Request

I want to add my project object detection which is made using yolov8 model developed by ultralytics .
It detects the objects which are been seen through the camera. It detects accurately.
Assign me this issue under GSSOC label as i am Gssoc'23 contributor

https://github.com/ayush9492

Define You

• GSSOC Participant
• Contributor

Project Name

Object Detection using Yolov8 model

Description

Project is based on object detection it can detect the object which are been seen through camera .
This model is trained using open source tool yolov8 which is been developed by ultralytics which helps in various face recognition, object detection , image specification tasks .
It can also be helpful to develop the models for fire detection, vehicle detection , etc

Project Request

Object detection for self-driving cars

Object detection for autonomous vehicles like self driving cars

Define You

• [Yes ] GSSOC Participant
• [ Yes] Contributor

Project Name

Object Detection for self driving cars

Description

Using YOLO models to detect objects such as Pedestrians, Vehicles, traffic signals, Lanes etc.

Timeline

Start date - Assigned day
End date - End of June

Project Request

Define You

• GSSOC Participant
• Contributor

AI Email Generator

Project Description

The AI Email Generator is a machine learning tool that aims to generate personalized emails using natural language processing techniques. The tool will use a dataset of email templates and user-specific information to create unique emails that sound human-written.

Scope

The scope of this project includes the development of an AI-based email generator tool that can automate the email writing process. The tool will be trained on a dataset of email templates, which will be used to generate emails based on user-specific information such as name, company, and event date. The tool will also be able to suggest appropriate email templates based on user input.

Deliverables

The project will deliver an AI-based email generator tool that can be used to write personalized emails. The tool will be deployed on Github and will include the following components:

• A web interface with good UI

Future Considerations

In the future, the tool can be extended to include more advanced natural language processing techniques such as sentiment analysis and topic modeling. The tool can also be integrated with popular email clients such as Gmail and Outlook.

Project Request

Detect parking slot occupancy in large parking spaces, thus saving fuel and time for the drivers.

https://github.com/PrathmeshN99

• GSSOC Participant
• Contributor

Project Name

PARKING SPOT OCCUPANCY DETECTION SYSTEM

Description

Finding a parking space nowadays becomes an issue, that is not to be neglected, it consumes time and energy. Using Computer Vision techniques and CNN classification, real-time car parking occupancy can be detected.
To develop an automated system to detect the empty parking spaces in a parking lot
To provide effective management of car parking to the users which can help in reducing the queues, improving scalability and the time required to find parking spaces

Block diagram

Project Request

The AI Image Search project is focused on developing an advanced system that utilizes artificial intelligence techniques to enable efficient and accurate searching of images based on their visual content.

Problem Statement:
The problem addressed by the AI Image Search project is the inefficiency and lack of accuracy in traditional image search methods. Conventional approaches rely on manually assigned metadata or textual descriptions, which can be subjective, incomplete, or time-consuming to generate. Additionally, searching for visually similar images or specific visual characteristics is challenging without an automated system capable of understanding and analyzing image content.

https://github.com/ayush-09

Define You

• GSSOC Participant
• Contributor

AI Image Search System

Description

The AI Image Search project is focused on developing an advanced system that utilizes artificial intelligence techniques to enable efficient and accurate searching of images based on their visual content. The project involves training a deep learning model on a diverse dataset of labeled images, extracting meaningful features from images, creating an index for storage and retrieval, and implementing a user-friendly interface for search queries and result presentation. The ultimate goal is to revolutionize image search in various domains, enhancing image retrieval, content organization, and user experiences.

Goals

• Develop an AI-powered image recognition model capable of accurately understanding and analyzing the visual content of images.
• Design a user-friendly interface that allows users to easily initiate image searches, provide reference images or keywords, and obtain relevant and visually similar results.
• Improve user experiences by providing a powerful and intuitive image search tool that saves time and effort in finding the desired images.

Expected Outcomes

• An end-to-end AI Image Search system that can perform accurate and efficient image searches based on visual content.
• Improved image retrieval capabilities, enabling users to find specific images or discover visually similar ones.
• Increased productivity in domains such as e-commerce, where product searches can be enhanced with visual-based queries.
• Enhanced user experiences with a user-friendly interface that simplifies the image search process.

Scope

The future scope of the AI Image Search project includes advancements in accuracy, fine-grained search, cross-modal search, contextual understanding, interactive user feedback, scalability, real-time processing, application integration, and exploring cross-domain applications.

Timeline

Start Date: When Assigned
End Date: 10 August

Project Description:
We are seeking assistance in creating a Jupyter notebook for training the YOLOv8 object detection model on a custom dataset. YOLOv8 is a state-of-the-art deep learning model known for its accuracy and real-time object detection capabilities. By training the model on our custom dataset, we aim to achieve accurate and efficient detection of specific objects relevant to our application.

Project Goals:

1. Develop a Jupyter notebook that facilitates training the YOLOv8 model on a custom dataset.
2. Enable the customization of hyperparameters, such as learning rate, batch size, and training epochs, to optimize model performance.
3. Provide support for data preprocessing, including annotation creation and bounding box adjustments.
4. Implement transfer learning from pre-trained weights to expedite training and improve convergence.
5. Incorporate data augmentation techniques to enhance the model's ability to generalize and handle variations in object appearance.
6. Integrate model evaluation metrics to assess detection performance, including precision, recall, and mean average precision (mAP).
7. Include instructions and documentation within the notebook to guide users through the entire training process, from data preparation to model evaluation.

Personality Prediction - From CV/Resume

A system that predict the personality of a person with O-C-E-A-N values from user's resume.

Atri Chatterjee

Define You

• [ ✔️ ] GSSOC Participant
• Contributor

Project Name

Description

The project has developed a system that utilizes various factors such as gender, age, openness score, conscientiousness score, extraversion score, agreeableness score, neuroticism score, and experience to make predictions about individuals' personalities. This system extracts relevant information from CVs/resumes and presents it on a result page. Logistic regression is employed to train the model, and the parsing of resume information is achieved using the pyresparser module, which is built using the nltk and spaCy modules in Python.

Scope

This system can be used in many business parts/areas that may require expert candidates. This system will reduce the workload of the Hiring Managers and help them (related to workers in general) to select the right candidate for the desired job profile, which in turn provides the expert (all the workers in a company or country) for the organization. Admin can easily shortlist a candidate based on their personality scores and select the appropriate candidate for a particular job profile

Timeline

[Start Date - When Assigned
End Date- 31 st July]

Project Request

House price detection

Define You

• GSSOC Participant
• Contributor

Project Name

Description

I want to build an AL training model, which an able learn from its previous fails and develop itself.
A good GUI ,where player will compete with AI model.
[Description of the project, its goals, and expected outcomes]

Scope

Timeline

10 June is end date.

Video Links or Support Links

[Links that can support the project in anyway]

Project Request

The objective of this project is to develop a spam alert system application that can detect and filter spam in calls, emails, and SMS using AI, machine learning, and deep learning techniques.

https://github.com/Ashgen12

Define You

• GSSOC Participant
• Contributor

Project Name

Spam Alert System Application using AI, Machine Learning, and Deep Learning Techniques

Description

-Objective: The objective of this project is to develop a spam alert system application that can detect and filter spam in calls, emails, and SMS using AI, machine learning, and deep learning techniques.

-Background: Spam is a major problem for communication service providers and users. It wastes time and resources and can even be harmful if it contains malicious content. Several machine learning and deep learning techniques have been used to detect and filter spam, including Naïve Bayes, decision trees, neural networks, and random forest.

-Methodology: The proposed spam alert system will use a combination of machine learning and deep learning techniques to detect and filter spam. A pre-trained transformer model like BERT (Bidirectional Encoder Representations from Transformers) can be fine-tuned to detect spam emails from non-spam (HAM). Similar techniques can be used to detect spam in calls and SMS.

-Expected Outcomes: The expected outcome of this project is a fully functional spam alert system application that can accurately detect and filter spam in calls, emails, and SMS.

Scope

-Research: Conduct research on the latest techniques and algorithms used to detect and filter spam in calls, emails, and SMS.

-Development: Develop a spam alert system application that uses AI, machine learning, and deep learning techniques to accurately detect and filter spam.

Timeline

Start Date: when assigned
End Date: 20 August

Project Request

I would like to propose a new project idea for developing an NLP-based text summarizer application that accepts PDF files as input and provides summarized content. Furthermore, the project aims to expand the functionality by implementing a content categorization feature to enhance the usability and organization of the application.

https://www.github.com/nayan-khemka

Define You

• GSSOC Participant
• Contributor

Project Name

Text Summarizer with PDF input and Content Categorization

Description

The objective of this project is to create a robust and user-friendly NLP Text Summarizer capable of handling PDF files directly. The application will extract text from PDFs, process and summarize the content using natural language processing techniques, and provide categorized summaries based on the content's topics or themes.

Scope

The project would only work with pdf with text as per what we are planning to do. It might extend it boundaries later in future.

Timeline

Start Date: Once the idea gets accepted and assigned
End Date: 10 days from Start

Video Links or Support Links

--

Project Request

This project involves using deep learning and computer vision techniques to develop an algorithm to identify metastatic cancer in small image patches obtained from larger digital pathology scans. The dataset using for this project will be comprising of Positive Cell Adenocarcinoma Margin (PCAM) images.

https://github.com/Anurag9492722884

Define You

• GSSOC Participant
• Contributor

Project Name

CanDetect

Description

-This project uses deep learning in PyTorch and computer vision techniques to develop an algorithm to identify metastatic cancer in small image patches obtained from larger digital pathology scans.

-The project leverages pre-trained Convolutional Neural Networks (CNNs) and transfer learning to improve the model's performance.

Scope

-Early Cancer Detection: One of the primary objectives of cancer image detection is to identify potential cancerous regions at an early stage. By utilizing deep learning models built with PyTorch, the project can help analyze medical images, such as early signs of cancerous growths or anomalies.

-Accurate Diagnosis: Deep learning models trained on large datasets can learn complex patterns and features that may be difficult for human experts to detect. By leveraging PyTorch's capabilities, the project can develop models capable of accurately diagnosing cancer.

-Research and Development: The project can contribute to the ongoing research and development efforts in the field of cancer diagnosis and treatment.

Timeline

Start time: Time of Assignment
End time: July 15th.

Project Request

Automatic captioning of videos based on cooking by understanding the action behind the scene and providing Nutrtional information based on the ingredients used.

https://github.com/nehavish006

Define You

GSSOC Participant
Contributor - Snekha C | Contributor

Project Name

Description

Video Captioning is a task of automatic captioning a video by understanding the action and event in the video which can help in the retrieval of the video efficiently through text. Video Captioning is an encoder decoder mode based on sequence to sequence learning. Automated video caption generator helps searching of videos in websites better and make content easier. The video information is considered as a sequence of images with 10 to 12 seconds short video clips. In the proposed system, Convolutional Neural Networks (VGG 16) is used for feature extraction and LSTM is used for encoding and decoding the features. Greedy search algorithm is used for predicting the efficient caption and gives speedy word extraction. Finally, the language converter is used for specific people language to understand the captions.

Scope

Objectives:

• To help the people with Deaf and hard of hearing individuals to watch videos helps people to focus on and remember the information more easily.
• To build the feature extraction using the convolutional neural network (VGG 16).
• To examine how LSTM is used for encoding and decoding the features.
• To explore how the greedy search algorithm predict the efficient caption.

Deliverables:

• A trained CNN RNN model capable of analyzing the video and generating the captions.
• Well-documented guidelines, including dataset preparation, training, inference, and any specific requirements.

Timeline

Start date : The date of assignment
End date: June 15 2023

Video Links or Support Links

Sandeep Samleti , Ashish Mishra , Alok Jhajhria , Shivam Kumar Rai, Gaurav Malik, 2021, "Real Time Video Captioning Using Deep Learning", International Journal Of Engineering Research & Technology (IJERT) Volume 10, Issue 12 (December 2021).

as you can see the navbar is not displayed properly so i want to work on this issue.

Project Request

To create a ML model that can detect objects in real time

https://github.com/lcs2022026?tab=overview&from=2023-05-01&to=2023-05-21

Define You

• [YES ] GSSOC'23 Participant
• [YES ] Contributor

@adithya-s-k
i would like to work on this project
can you please assign it to me

Project Request

A project which classifies the genre of a music file and uses a gradio based web GUI for the same

https://www.github.com/infernalsaber

Define You

• GSSOC Participant
• Contributor

This idea has been approved

Project Name

MelodyNet

Description

Using transfer learning and audio preprocessing, we aim to classify the genre of a song

Scope

This can be used to classify old music files that a person possesses. Also useful for artists to label their song to reach their desired audience

Timeline

Start date: when assigned
End date: 30 June

#Links

Adil Lheureux, Music genre classification

name: Update Request
about: If you want to make any updates to a project
title: "[UPDATE]"
labels: ''
assignees: ''

https://github.com/avijit1999

Define You

• GSSOC Participant
• Contributor

Is your feature request related to a problem? Please describe.
Update the readme of artifitial_inteligence. add project name and contributer

Project Request

Developing a model to predict early heart disease prediction by doing a comparitive analysis of different machine learning models to predcit heart disease with highest accuracy on defined and most important attributes

https://github.com/poorvika11

Define You

• GSSOC Participant
• Contributor

Early Heart Disease Prediction

Description

The project involved analysis of the heart disease patient dataset with proper data processing. Then, different models will be trained and predictions will be made with different algorithms KNN, Decision Tree, Random Forest,SVM,Logistic Regression etc. Variety of Machine Learning algorithms implemented in Python to predict the presence of heart disease in a patient. This is a classification problem, with input features as a variety of parameters, and the target variable as a binary variable, predicting whether heart disease is present or not.

Scope

Objectives:

• Implement a robust and efficient heart disease prediction.
• Enable high accuracy model using comparitive analysisof all machine learning models.

Deliverables

• A trained model capable of predciting disease accurately.
• Well-documented guidelines, including dataset preparation, training, inference,exploratory data analysis and any specific requirements.
• Models predicting using different algorithms KNN, Decision Tree, Random Forest,SVM,Logistic Regression etc.

Timeline

Start Date: when assigned
End Date: 10 August

Update Request: Adding Error Handling and Validation to the Code
After reviewing the existing code, I have identified the need for incorporating error handling and validation mechanisms. These additions will help improve the reliability and usability of the application, ensuring that users have a seamless experience. Here are the specific changes I propose:

OpenAI API Key Validation:

Implement a validation function to verify the format and validity of the provided OpenAI API key.
Display a warning message if an invalid or empty API key is entered.
Email Input Validation:

Introduce a validation function to check the length of the email input.
Set a maximum word limit (e.g., 700 words) and display an error message if the input exceeds the limit.
Error Handling:

Handle potential errors during API requests to OpenAI and display meaningful error messages to users.
Gracefully handle any unforeseen errors or exceptions to prevent the application from crashing or behaving unexpectedly.
By incorporating these changes, the application will provide a more robust and user-friendly experience, ensuring that users can generate accurate and formatted emails without encountering issues or unexpected errors.

I am willing to contribute to this update by working on the code changes and submitting a pull request. However, I would appreciate your guidance and feedback on the proposed approach before proceeding.