Santa Clara, California, United States
3K followers 500+ connections

Join to view profile

About

Building Large Vision-Language and Foundation Models toward Artificial General…

Activity

Join now to see all activity

Experience & Education

  • NVIDIA

View Subhashree’s full experience

See their title, tenure and more.

or

By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.

Licenses & Certifications

Volunteer Experience

  • Make A Difference Graphic

    ED-Support Volunteer

    Make A Difference

    - 2 years 4 months

    Education

    Provided educational assistance to Institutionalized children weekly once

  • AI4ALL Graphic

    Mentor

    AI4ALL

    - Present 4 years 8 months

    Education

    Changemakers in AI : Mentoring high school students working on impactful AI projects

Publications

Patents

Courses

  • Advanced Computer Vision

    -

  • Advanced Machine Learning

    -

  • Advanced Microcontrollers

    371

  • Artificial Intelligence

    -

  • Calculus, Matrix Algebra and Ordinary Differential Equations

    111

  • Computer Programming

    100

  • Computer Vision

    -

  • Deep Learning

    CS5984

  • Digital Signal Processing

    221

  • Digital Systems

    210

  • Electrical Measurements and Instrumentation

    213

  • Embedded Systems Design

    471

  • Engineering Drawing

    181

  • Introduction to Microcontrollers and Applications

    310

  • Mathematical Statistics and Numerical Methods

    212

  • Measurements and Analog Circuit Lab

    392

  • Signals and Systems

    220

  • Stochastic Signals and Systems

    -

Projects

  • Performance Analysis of Different Techniques for Fight Detection in Videos

    Different Feature Extraction Methods and Classification Methods were implemented and their performances were compared:
    1.STIP Feature Extraction with SVM Classifier
    2.Augmented Affine Trajecton with SVM Classifier
    3.CNN Classifier
    4.CNN feature extraction with SVM, KNN, and Decision Tree Classifiers

    Other creators
    See project

  • Object Localization in Images Using Referral Expressions

    -

    1. I had implemented a multi-modal embedding using Multi Layer Perceptron to map corresponding location and referral expressions. The bounding box was obtained using SSD during testing phase.
    2. Different text feature extraction methods such as skip-thought, GLOVE and word2vec was experimented and compared.
    3. Losses such as contrastive loss, L2 loss. The highest recall was obtained for skip-thought and order embedding.

    See project

  • Semi-Supervised Domain Adaptation Using CycleGAN [PyTorch]

    -

    - Proposed New architecture for Domain adaptation for the task of semantic segmentation.
    - The proposed architecture used combined Cyclic GAN and FCN where cyclic GAN was a better regularizer than Domain Adversarial Training and was designed to perfrom domain adaptation between source target and partially labeled target domain.
    - The entire implementation was done in PyTorch.

    See project

  • Advanced Computer Vision - Course [Matlab]

    -


    1. EdgeDetection_HybridImgs-This project involves forming Hybrid Images and the implementation of Canny Edge Detection without inbuilt function.
    2. EpiPolarGeometry_Metrology_StructureFromMotion- This project is about finding the height of building from single view metrology, Predicting the motion of keypoints using optical flow
    3.DescrptorMatching_KeyPointTracking_objectInstanceRecognition- This project involves keypoint tracking, Shape alignment of one with another given two images…


    1. EdgeDetection_HybridImgs-This project involves forming Hybrid Images and the implementation of Canny Edge Detection without inbuilt function.
    2. EpiPolarGeometry_Metrology_StructureFromMotion- This project is about finding the height of building from single view metrology, Predicting the motion of keypoints using optical flow
    3.DescrptorMatching_KeyPointTracking_objectInstanceRecognition- This project involves keypoint tracking, Shape alignment of one with another given two images and object instance recognition
    4.SLIC_GraphCutSegmentation_EMAlgorithm- This project involves implementation of SLIC and Grab Cut algorithms for segmentation and implementing EM algorithm for finding mean scores of annotated images.
    5.SceneCategoryRecognition_BagOfWords_differentClassifiers- This project involves outdoor category classification using Bag of Word Model and SVM, KNN classifiers.

    See project

Honors & Awards

  • Summer Research Fellowship

    Indian Academy of Sciences

    Was one of the 80 fellows selected out of 2000 candidates across India.

  • Amrita Best Public Speaker Award

    Centre for Corporate and Industry Relations (CIR), Amrita Vishwa Vidyapeetham

  • Outstanding Student Award 2014-2015

    IEEE-IAS

    Awarded to one student every year based on his/her excellence in Academics and Extra-curriculars

  • Academic Proficiency Award

    Amrita Vishwa Vidya Peetham

    2013-2015

    Awarded for topping the department based on Academic credentials.

Languages

  • English

    Full professional proficiency

  • Tamil

    Native or bilingual proficiency

  • German

    Limited working proficiency

  • Hindi

    Professional working proficiency

Organizations

  • IEEE Industrial Applications Society

    Student Member

    - Present

More activity by Subhashree

View Subhashree’s full profile

  • See who you know in common
  • Get introduced
  • Contact Subhashree directly
Join to view full profile

Other similar profiles

Explore top content on LinkedIn

Find curated posts and insights for relevant topics all in one place.

View top content

Others named Subhashree Radhakrishnan

1 other named Subhashree Radhakrishnan is on LinkedIn

See others named Subhashree Radhakrishnan

Add new skills with these courses

See all courses