PeopleofSuperintelligence

Researcher

Rohit Girdhar

Rohit Girdhar is a computer vision researcher whose work spans video understanding, multimodal representation and media generation. He earned a computer science degree at IIIT Hyderabad, completed a robotics PhD at Carnegie Mellon, and joined Meta FAIR in 2019. He coauthored the Anticipative Video Transformer, Omnivore, ImageBind, Emu Video and Movie Gen, spanning action prediction, shared visual representations and video generation. He has been a research scientist at AMI Labs since June 2026, where he works on world models; he was named to MIT Technology Review’s 2026 35 under 35 Asia Pacific list.

Now at
AMI Labs
Works
53
Citations
9,890
h-index
28
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2010–2019

    Degrees at IIIT and CMU

    He earns a computer science BTech at IIIT Hyderabad in 2014 and completes CMU’s MSR program in May 2016. At Carnegie Mellon University’s Robotics Institute, he completes a robotics PhD in August 2019 under advisor Deva Kannan Ramanan; his thesis studies people through local, global and temporal reasoning.

  2. 2016–2018

    Research internships

    During 2016–2018, he works as a research scientist intern at Adobe Research, Facebook AI Research and DeepMind on action classification, video pose tracking and video action recognition.

  3. 2019–2021

    Research at Meta FAIR

    In 2019, he joins Meta’s Facebook AI Research (FAIR) as a research scientist, studying self-supervised, multimodal and omnivorous representations for video understanding. With Kristen Grauman, he publishes the Anticipative Video Transformer in 2021, an attention-based model that anticipates future actions from observed video.

  4. 2022

    Omnivore across visual modalities

    In 2022, he coauthors Omnivore, a single model that classifies images, videos and single-view 3D data with the same model parameters.

  5. 2023

    ImageBind and six modalities

    He coauthors ImageBind in 2023, which learns a joint embedding across images, text, audio, depth, thermal and IMU data.

  6. 2023–2024

    Emu Video and Movie Gen

    In 2023, he coauthors Emu Video, which generates an image from text before using it to condition video generation. In 2024, he coauthors Movie Gen, a family of foundation models that generates 1080p video with synchronized audio.

  7. 2026

    MIT Technology Review TR35

    He receives MIT Technology Review’s 35 under 35 (TR35) Asia Pacific recognition in 2026.

  8. Since June 2026

    Research scientist at AMI Labs

    He joins AMI Labs as a research scientist in June 2026 and works on multimodal understanding, generation and world modeling.

Career

  1. Since 2026AMI LabsResearch scientist
  2. 2020–2024Meta
  3. 2016–2018AdobeResearch Scientist Intern
  4. 2016–2021FAIR (Meta)Research Scientist Intern, Research Scientist
  5. 2016–2018Google DeepMindResearch Scientist Intern
  6. 2015–2017International Institute of Information Technology, Hyderabad

Education

  1. 2019Carnegie Mellon UniversityPhD
  2. 2016Carnegie Mellon UniversityMSR
  3. 2014International Institute of Information Technology, HyderabadBTech

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

Most-cited works

  1. Masked-attention Mask Transformer for Universal Image Segmentation2022 · IEEE Computer Society Conference on Computer Vision and Pattern Recognition · 3,513 citations
  2. ImageBind One Embedding Space to Bind Them All2023 · IEEE Computer Society Conference on Computer Vision and Pattern Recognition · 857 citationsWith Zhuang Liu
  3. Learning a Predictable and Generative Vector Representation for Objects2016 · Lecture notes in computer science · 673 citationsWith Abhinav Gupta
  4. Ego4D: Around the World in 3,000 Hours of Egocentric Video2022 · IEEE Computer Society Conference on Computer Vision and Pattern Recognition · 626 citationsWith Antonio Torralba, Aude Oliva, C. V. Jawahar, Christoph Feichtenhofer, Dhruv Batra and Dima Damen
  5. Detecting Twenty-Thousand Classes Using Image-Level Supervision2022 · Lecture notes in computer science · 559 citations

From OpenAlex, as of 21 October 2026.

  1. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes Rohit Girdhar. Our estimates of where the 943 people shaping AI who worked at Google and Google DeepMind work now: 446 at Google and 452 across 278 other organizations.
  2. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Rohit Girdhar. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  3. Google had the most people shaping AI every year from 2014 to 2025Includes Google. Our estimates of which organizations employed the most people shaping AI each year from 1955 to 2025, tracing the shift from university labs to companies.
  4. Google has the highest combined h-index among AI organizationsIncludes Google. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.

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 Rohit Girdhar yet. Send us a free-licensed photo.

  1. Rohit Girdhar rohitgirdhar.github.io. Checked 20 October 2026.
  2. Rohit Girdhar - Robotics Institute Carnegie Mellon University ri.cmu.edu. Checked 20 October 2026.
  3. Anticipative Video Transformer arxiv.org. Checked 20 October 2026.
  4. Omnivore: A Single Model for Many Visual Modalities arxiv.org. Checked 20 October 2026.
  5. ImageBind: One Embedding Space To Bind Them All arxiv.org. Checked 20 October 2026.
  6. Emu Video: Factorizing Text-to-Video Generation by Explicit Image Conditioning arxiv.org. Checked 20 October 2026.
  7. Movie Gen: A Cast of Media Foundation Models arxiv.org. Checked 20 October 2026.