IamtheFounderandChiefScientistofFastCodeAIandholdaPh.D.inComputerSciencefromtheMax-PlanckInstituteforInformaticsinGermany.Myresearchintersectscomputergraphics,vision,andmachinelearning.IalsoserveasanAdjunctFacultymemberintheCDSdepartmentatIISc,whereIleadadeeplearningresearchgroup.Withanextensiveteachingandprofessionalbackground,IhavetaughtComputerVisionatIITBombayandworkedwithcompanieslikeYahoo!,WetaDigital,andApple.
PhD from The Max Planck Institute for Informatics
Graduated Summa Cum Laude, PhD advisor: Prof. Dr. Thorsten Thormaehlen, Prof. Dr. Hans-Peter Seidel,
Thesis: Data-driven Methods for Interactive Visual Content Creation and Manipulation Link Download
Graduated from University of Saarland
Bachelor of Engineering in Computer Science from R.V. College of Engineering, Bangalore, India.
Rishabh Dabral
Rishabh Dabral
Rahul Mitra
Rahul Mitra
Nehal Doiphode
Nehal Doiphode
Uday Kasupati
Uday Kasupati
Shubham Goel
Shubham Goel
Saurabh Sharma
Saurabh Sharma
Anurag Mundahada
Anurag Mundahada
Nitesh Bharadwaj Gundavarapu
Nitesh Bharadwaj Gundavarapu
Naman Jain
Naman Jain
Rohit Jena
Rohit Jena
Sahil Shah
Sahil Shah
Pavan Teja
Pavan Teja
Divyansh
Divyansh
Jai Doshi
Jai Doshi
Sept 2015 – Present
Founder and CEO
Founder and CEO, Fast Code AI Consult Pvt. Ltd., Bangalore, India
Jan 2020 – Present
Adjunct Faculty
Adjunct Faculty @ CDS Dept., IISc
Deep Learning for Computer Vision (Spring 2020 & Spring 2021), with Prof. R. Venkatesh Babu
Dec 2016 – May 2019
Adjunct Assistant Professor
Adjunct Assistant Professor @ CSE Dept., IIT Bombay
Computer Vision (Spring 2018 & Spring 2019)
Deep Learning module for Computer Vision (Spring 2017)
July 2015 – Nov 2015
Senior Researcher
Senior Researcher at SPG, Apple, Cupertino
March 2013 – Feb 2015
Post-Doc
Post-doc at Courant Institute, NYU with Chris Bregler and Yann LeCun
Feb 2011 – May 2011
Research Intern
Research Intern at Weta Digital Wellington, New Zealand.
Feature film credits: The Adventures of Tintin(2011) – R & D Developer
Oct 2007 – Aug 2008
Research Assistant
Research Assistant at Media Integration and Communication Center (MICC), University of Florence, Florence, Italy
Sept 2006 – Sept 2007
Software Developer
Software Developer at Yahoo India R & D, Bangalore, India
Embodiments of present disclosure relates to method for identifying a hand pose in a vehicle, and a system for performing an event in a vehicle. A hand image for a hand in the vehicle, is extracted from a vehicle image of the vehicle for identification. Plurality of contextual images of the hand image is obtained based on the single point. A hand pose associated with the hand is identified based on the plurality of contextual features using a classifier model.
Advanced airbag deployment control system designed for vehicular use incorporating an image sensor unit that captures real-time images of a vehicle occupant, focusing particularly on head localization. Utilizing a processing unit equipped with Long Short Term Memory (LSTM) neural network architecture, the system analyzes images to determine and predict the future position and orientation of the occupant's head. The system dynamically adjusts the direction in which the airbag flap is removed and the airbag's inflation pressure, ensuring optimal safety by adapting the deployment to the predicted head position at the moment of impact.
A method for activating vehicle functions via hand gestures, utilizing a 3D Convolutional Neural Network (3D-CNN) and Gated Recurrent Unit (GRU) to analyze video frames and extract spatio-temporal features. A prediction module simultaneously assesses the gesture's progression and classifies it, employing predefined models to determine the gesture's type and its completion rate. Upon accurate detection and classification, a corresponding event is triggered within the vehicle.
The invention concerns a computer-implemented method for tracking and reshaping a human-shaped figure in a digital video comprising the steps: acquiring a body model of the figure from the digital video, adapting a shape of the body model, modifying frames of the digital video, based on the adapted body model and outputting the digital video.