Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings
@sanskaromar
sanskaromar
Follow
View sanskaromar's full-sized avatar
:octocat:

Sanskar Omar sanskaromar

:octocat:

Organizations

@ContriHUB@potehtoT@HopeChainHQ@gdsc-mnnita

Block or report sanskaromar

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more aboutblocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more aboutreporting abuse.

Report abuse
sanskaromar/README.md

sanskaromar

Hi I'm Sanskar!

A passionate developer from India

Typing SVG

  • 🏫 I’ve completed my B.Tech inComputer Science & Engineering from MNNIT Allahabad in June 2024.
  • 🏢 I'm currently a Software Engineer at VISA.
  • Check out my portfolio athttps://sanskaromar.github.io/me/

  • 👨‍💻 All of my projects are availablehere
  • ✉️ Reach out to me athey[My first Name]0[dot]0[at]gmail[dot]com
  • 😁 Fun fact :I love Anime and Cats

ProjectTech StackDescriptionLink
Tensor Fusion based Multi-modal Transformer-model for Depression DetectionPython, Transformer Models, Tensorflow
  • Developed pipeline for extracting facial landmark features and low-level descriptors (LLDs) from videos to generate visual and acoustic encodings for modeling.
  • Proposed a novel transformer architecture incorporating self-attention for independent feature extraction, cross-attention for multimodal fusion, and tensor fusion layer to model semantically distinct subregions capturing unimodal and bimodal dynamics.
  • Attained an F1-score of 67.02% - an improvement over the SOTA model utilizing a robust model architecture.
Paper under Review
Medical Sound Classification ChallengeAudio Classification, Embeddings processingBuilt models to classify respiratory diseases like asthma and COPD using audio files and medical features for accurate diagnosisleaderboard
Frontline ConnectAndroid(Kotlin), Firebase, React, Render, Netlify, Node JS,A geofencing location-based attendance management system leveraging biometric authentication and real-time tracking to streamline frontline duty personnel (such as policemen, health workers and volunteers) management atPrayag Kumbh Mela built using Android (Kotlin), React.js, Firebase, and Node.js to replace traditional manual error-prone tracking methods.B.Tech Major project so code/report not public
Realtime Posture Detection and Alert System for Desk Workers using Wearable devicesPython, Tensorflow-lite, Raspberry Pi 4B, Java, Android,Developed an integrated posture monitoring system that combines computer vision-based real-time posture detection with a mobile app interface for users and cloud storage, implemented on a low compute embedded device with accurate posture analysis (accuracy score ~0.92 and realtime alerts, user feedback, leveraging deep learning for precise posture correction.paper
Avishkar 2k22PhasorJS, ReactJs, NodeJs, PlanetScale, SQLA website for annual tech fest of MNNIT Allahabad. A 2D top-down RPG with the campus as the map and quizzes, registration management, and leaderboardsitecode
HopeChainBlockchain, Solidity, React, Tailwind CSS, Kotlin, AndroidX, Jetpack-Compose, Polygon, Etherium, Hardhat, Alchemy, Remix, MetamaskA decentralized donations management platform to tackle the world's most intimidating disasters by providing funds quickly and transparently.democode
cWuzzleNextJs, FastAPI, SupaBaseA website about crossword puzzles. Create, Play, and Share fun crosswords on the go with the latest features like autosave, clue highlight, private sharing, crossword regeneration, and many more...cWuzzle,devfolio
GestoMediapipe, Tensorflow.js, NextJs, FastAPI, FirebaseA first of its kind web app for custom gesture training and recognition with Zero Prediction Latency, Faster Training, and High Accuracy, a startup that classified sign language gestures. selected for Google Startup Bootcamp 2022 and top 30 startups at Nasscom Cisco Thinqbator Cohort 5here
CertifaiPythonGenerate bulk custom certificates from template superfast and with ease⚡, Built this for Google Developer Student Club's Communityhere
Dr. VincentUnity3D, C#A RPG game based around theme of covid-19 pandemiccode
CollabarushNextJS 13, SupaBase, SocketIOA real time collaborative paint applicationdemocode
Logical Rhythm 2k21Classification, Regression, Image ClassificationWon Logical Rhythm 2k21, which included these Kaggle competitions followed by a presentation round where we defend our approach based on EDABeans,Gganbu?,Cars24
Logical Rhythm 2k22Classification, Regression, Image ClassificationWon Logical Rhythm 2k22, which had similar format as LR 2k21Finding Disney,Motorbike Cost,Patient Healthcare
Logical Rhythm 2k23 - Kaggle Competitions HostNatural Language Processing, Basic Classification and Regression based tasksHosted Logical Rhythm 2k23 - An annual machine learning event at NIT Allahabad focused on encouraging Data Science and ML culture in collegePilot,Detective Bondman,Space shop,The Aerospace Enigma
SecureFacePython, OpenCV, UnittestAn open-source project designed to provide facial recognition-based access control for your applications, used in a Hacktoberfest-like college event to teach juniors about open-source contributions and help them get started with ithere,open-source fork
Track My TasksReactMinimalistic Personal Tasks Tracker built with open source contribution from various devshere
Culrav 2k22HTML, CSS, JS, PhPCulrav is the annual cultural fest of MNNIT Allahabad. I was part of a two-member web team for Culrav 2k22.here
protectHerReactA woman safety app and police clientuser codepolice codeuser ClientPolice Client
E-cell WebsiteReact, Node JSWebsite for annual event of Ecell MNNITcode
Calculator with quotesReactA simple calculator and quote display apphere
Country AppReactDisplay data from rest country apihere
Mood Detection and Song RecommendationPython, ML, DLIt does what the name suggestshere

My Social Profiles :

https://www.linkedin.com/in/sanskaromar-/sanskaromar_sanskaromar7

PinnedLoading

  1. certifaicertifaiPublic

    Certificate Generator | Generate bulk custom certificates from template superfast and with ease⚡

    Python 11 2

  2. Dr.VincentDr.VincentPublic

    The winning project of WoC 2021

    3 2

  3. DuP-491/AvishkarDuP-491/AvishkarPublic

    TypeScript 2 2

  4. ContriHUB/SecureFaceContriHUB/SecureFacePublic

    Forked fromsanskaromar/SecureFace

    SecureFace Access Control System

    Python 4

  5. en1gm479/Song_Recommendation_using_Facial_Expressionen1gm479/Song_Recommendation_using_Facial_ExpressionPublic

    This Project was developed as a part of Robotics Club Summer Projects.

    Python 1

  6. renaissance-2k23renaissance-2k23Public

    This website is under development for Renaissance 2023 - Annual event by E-cell MNNIT

    JavaScript 3 13


[8]ページ先頭

©2009-2025 Movatter.jp