PeopleofSuperintelligence

Researcher

Arun Balajee Vasudevan

Arun Balajee Vasudevan is a computer vision researcher whose work connects language, vision, sound and autonomous navigation. He earned degrees at IIT Jodhpur and EPFL, completed a PhD at ETH Zurich, and held a postdoctoral role at Carnegie Mellon. His papers span query-guided video summaries, gaze-aware object referring, binaural sound learning and world models for autonomous driving. He was a Research Scientist at Amazon in June 2025 and is an Assistant Professor in IIT Madras's DSAI department as of October 2026.

Badges

Timeline

  1. 2014–2016

    Degrees at IIT Jodhpur and EPFL

    He earns a bachelor's in electrical engineering at IIT Jodhpur in May 2014 and a master's in computer science at EPFL in September 2016.

  2. 2017

    Query-adaptive video summaries

    With Michael Gygli, Anna Volokitin and Luc Van Gool, he coauthors an ACM Multimedia paper in 2017 that selects diverse, representative video frames relevant to text queries and introduces a dataset with diversity and query-specific relevance labels.

  3. 2018

    Language and gaze in videos

    With Dengxin Dai and Luc Van Gool, he coauthors a CVPR paper in 2018 that combines language, human gaze, appearance, motion and spatio-temporal context to locate referred objects in video.

  4. 2020–2022

    Binaural sound and visual learning

    In 2020, he coauthors a method for dense semantic labels of sound-making objects from binaural audio and records an audio-visual street-scene dataset. In 2022, he coauthors a study of self-supervised representations for binaural sound and image data using multiple pretraining tasks.

  5. 2024–2025

    Adaptive world models for driving

    With Neehar Peri, Jeff Schneider and Deva Ramanan, he develops BehaviorNet to predict reactive agent behavior and AdaptiveDriver to plan with learned world models in 2024. In 2025, their paper is accepted to ICRA and evaluates AdaptiveDriver on the nuPlan closed-loop planning benchmark.

  6. By March 2025

    An ETH PhD and CMU postdoc

    At ETH Zurich, he completes a PhD in the Computer Vision Lab under Luc Van Gool and Dengxin Dai before working as a postdoctoral fellow in Deva Ramanan's Robotics Institute lab at Carnegie Mellon. His work covers computer vision, multimodal perception and navigation, 3D/4D reconstruction, motion planning, and foundation models.

  7. June 2025

    Research Scientist at Amazon

    He completes his Carnegie Mellon postdoctoral role in March 2025 and is a Research Scientist at Amazon in June 2025.

  8. As of October 2026

    Assistant Professor at IIT Madras

    As of October 2026, he is an Assistant Professor in the DSAI department at IIT Madras.

Career

  1. NowIIT MadrasAssistant Professor
  2. Until 2025Carnegie Mellon University
  3. From 2025AmazonResearch Scientist

Education

  1. 2016EPFLMaster's degree
  2. 2014Indian Institute of Technology JodhpurBachelor's degree
  3. ETH ZurichPhD

From the biography.

  1. 23% of people shaping AI who left a university for a company later returned to a universityIncludes Arun Balajee Vasudevan. We estimate that 468 of the 2,011 people shaping AI who left a university for a company (23%) later returned to a university after a median of 5 years, and 163 of 829 (20%) made the reverse round trip.
  2. 16% of AI career moves in the 2020s went from a university to a companyIncludes Arun Balajee Vasudevan. We estimate that 16% of AI career moves in the 2020s went from a university to a company; 29% of moves starting at a university went to a company, and 76% ended the university role.
  3. Among employers with 30+ full-time cases, OpenAI leads primary technical role H-1B pay at $310,000, and Amazon, Google and Microsoft file 57% of all casesIncludes Amazon. Across 440,669 certified US H-1B filings from 44 employers (FY2020–FY2026), OpenAI leads primary technical role base pay at $310,000 and xAI leads overall base pay at $290,000.
  4. The United States employs an estimated 48% of people shaping AIIncludes IIT Madras. We estimate where the people shaping AI work today, how universities, companies and research institutes split across countries, and how country shares have changed since the 1960s.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Arun Balajee Vasudevan yet. Send us a free-licensed photo.

  1. Arun Balajee Vasudevan arunbalajeev.github.io. Checked 21 October 2026.
  2. [1705.00581] Query-adaptive Video Summarization via Quality-aware Relevance Estimation arxiv.org. Checked 21 October 2026.
  3. [1801.01582] Object Referring in Videos with Language and Human Gaze arxiv.org. Checked 21 October 2026.
  4. [2003.04210] Semantic Object Prediction and Spatial Sound Super-Resolution with Binaural Sounds arxiv.org. Checked 21 October 2026.
  5. [2201.01046] Sound and Visual Representation Learning with Multiple Pretraining Tasks arxiv.org. Checked 21 October 2026.
  6. [2406.10714] Planning with Adaptive World Models for Autonomous Driving arxiv.org. Checked 21 October 2026.
  7. Team - Uncertainty in Image classification mscvprojects.ri.cmu.edu. Checked 21 October 2026.
  8. Multimodal Learning in 3D environments: Perception, and Simulation homecse.iitd.ac.in. Checked 21 October 2026.