BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:Asia/Kolkata
X-WR-TIMEZONE:Asia/Kolkata
BEGIN:VEVENT
UID:103@cds.iisc.ac.in
DTSTART;TZID=Asia/Kolkata:20250204T113000
DTEND;TZID=Asia/Kolkata:20250204T123000
DTSTAMP:20250128T131326Z
URL:https://cds.iisc.ac.in/events/seminar-cds-102-february-04th-1130-multi
 modal-learning-in-3d-environments-perception-and-simulation/
SUMMARY:{Seminar} @ CDS: #102\, February 04th\, 11:30: "Multimodal Learning
  in 3D environments: Perception\, and Simulation"
DESCRIPTION:Department of Computational and Data Sciences\nDepartment Semin
 ar\n\n\n\nSpeaker :Arun Balajee Vasudevan\, postdoctoral researcher at Car
 negie Mellon University\nTitle :"Multimodal Learning in 3D environments: P
 erception\, and Simulation"\nDate &amp\; Time : February 04\, 2025 (Tuesda
 y)\, 11:30 AM\nVenue : # 102\, CDS Seminar Hall\n\n\n\nABSTRACT\nAutonomou
 s robots have several potential applications such as virtual assistants\, 
 VR/AR\, gaming\, self-driving technologies\, city planning and others. To 
 achieve autonomy\, a robot needs to see and hear the environment\, before 
 it can converse or navigate favorably to perform a human-desired task. Pre
 cisely\, scene understanding begins with building 3D geometry\, decoding s
 emantics\, understanding surround objects/humans\, and planning and action
 s. My research talk addresses these fundamental challenges independently u
 nder two broad themes: understanding geometry and multimodal perception &a
 mp\; data-driven simulation and navigation.\nUnder geometry and perception
 \, I introduce the usage of several multimodalities such as the user's gaz
 e\, visual sensors such as cameras or range sensors (e.g. Kinect)\, and sp
 eech/language instructions from human referrals for robot perception tasks
 . Further\, I delve deep into the investigation of the audio sensing modal
 ity for the task using binaural sound microphones. Secondly\, regarding th
 e geometry\, my talk addresses one of my ongoing works about the construct
 ion of digital twins of the real world with 4D reconstruction of dynamic s
 cenes from ground visuals of a robot.\nUnder the theme of Data-driven simu
 lation and robot navigation. Following perception\, robots must navigate a
 nd take meaningful actions in the world. This involves broadly two aspects
 : wayfinding and motion planning. Earlier works address wayfinding based o
 n directional instructions\, overlooking human aspects. I briefly talk abo
 ut a new paradigm that integrates principles from cognitive science with l
 earning-based methods to tackle the challenge of language-based wayfinding
  for robots in real-world outdoor environments. The second aspect is motio
 n planning for which I propose the learning of driver behavior models for 
 MPC-based planners to build data-driven simulators.\nLong term\, I envisio
 n to bridge the above two themes to build multimodal digital twins simulat
 ors of the real-world. This potentially helps in training and testing of p
 lanners\, VR/AR setups\, gaming\, and others. Lastly\, I also cover my fut
 ure research plan in the talk.\n\nBIO: Arun Balajee Vasudevan is currently
  a postdoctoral researcher at Carnegie Mellon University under Prof. Deva 
 Ramanan. His core research interest is in Computer Vision and Multimodal L
 earning. He has works in multimodal (vision\, language and sounds) percept
 ion and navigation\, 3D/4D reconstruction\, Motion Planning and improving 
 Foundational models. He published papers predominantly in vision and machi
 ne learning conferences/journals such as CVPR\, ECCV\, ICML\, IJCV\, TPAMI
 \, and others. He defended his PhD under Prof. Luc Van Gool at ETH Zurich.
  He received his MSc in Computer Science from EPFL in 2016 and an undergra
 duate degree in Electrical Engineering from the Indian Institute of Techno
 logy Jodhpur in the year 2014. From March 2025\, he would be joining Amazo
 n as a Research Scientist.\n\nHost Faculty: Dr. Anirban Chakraborty\n\n\n\
 nALL ARE WELCOME
CATEGORIES:Events,Talks
END:VEVENT
BEGIN:VTIMEZONE
TZID:Asia/Kolkata
X-LIC-LOCATION:Asia/Kolkata
BEGIN:STANDARD
DTSTART:20240205T113000
TZOFFSETFROM:+0530
TZOFFSETTO:+0530
TZNAME:IST
END:STANDARD
END:VTIMEZONE
END:VCALENDAR