Nikhil Shendge's Projects
Perform human pose estimation using Lightweight Human Pose Estimation.
Created the books-library using the OOPS method
This repository contains a Python script that implements object detection and tracking using the YOLO (You Only Look Once) algorithm. It utilizes the ultralytics library for YOLO object detection, OpenCV (cv2) for image processing, cvzone for graphical overlays, and the SORT (Simple Online and Realtime Tracking) library for object tracking.
Convolutional Neural Networks (CNN) form the basis of all image processing. It is also applied in Face Recognition
As you guys can see I used PIXELLIB and ALTER_BG'S XCEPTION_PASCALVOC.PG model to change the image background to white.
Chrome extension that leverages the latest GPT (Generative Pre-trained Transformer) techniques. The extension provides users with multiple options to interact with web pages, including summarization, paraphrasing, and content analysis.
This project is about web scrapping the lichess online chess gaming website user name kingscrusher's all old games and also the chess movies as well in the pgn format.
A Python script that leverages the power of the langchain package to create a CSV agent. This agent can answer questions about a CSV file using the OpenAI API. The code provides a user-friendly interface built with Streamlit, allowing users to upload a CSV file and inquire about its content.
Used CNN to detect if the person is wearing a mask or not
The code is a Python implementation of real-time face detection using OpenCV and Mediapipe libraries. It utilizes the computer vision capabilities of OpenCV and the face detection module from the Mediapipe library to detect faces in a live video feed or webcam.
The face mesh is a 3D model of a face. It works in combination with the face tracker in Meta Spark Studio to create a surface that reconstructs someone's expressions. Once you've added a face tracker and face mesh to your project you can create face tracking effects, add retouching or change the shape of the face.
This Python code demonstrates hand detection and tracking using the Mediapipe library and OpenCV. It utilizes the HandDetector class to detect and track multiple hands in a video stream. The code provides the positions of landmarks on the detected hands.
The handwritten digit recognition is the ability of computers to recognize human handwritten digits. It is a hard task for the machine because handwritten digits are not perfect and can be made with many different flavors.
EDA
From the given "Iris" dataset, predict the optimum number of clusters and represent it visually
Langchain with OpenAI API is a Python project that integrates the OpenAI API with the Langchain framework to generate information about celebrities based on user queries. It utilizes the Streamlit framework to create an interactive user interface and implements conversation buffer memories to store chat histories.
This is langchain chatbot using open-ai and huggingface embedded architect
This code is a Python script designed to facilitate question-answering on PDF documents. By leveraging libraries such as PyPDF2, streamlit, and langchain, it allows users to upload a PDF file, extract its text, split it into chunks, and create embeddings.
The system extracts text from uploaded PDFs, creates a vector store of the text chunks, and uses a conversational chain model to generate relevant answers based on user queries. It offers a user-friendly interface powered by Streamlit, making it easy to upload PDFs, ask questions, and receive responses in a conversational format.
my personal portfolio
RAG Python Solution with LLAMA3, LangChain, Ollama and ChromaDB in a Flask based solution.
This project is about web scrapping the olx website and getting their product image urls, product prices, product names and their types as well.