PeopleofSuperintelligence

Researcher

George Toderici

Also known as George D. Toderici

George Toderici is a computer vision researcher who worked on video understanding and neural image compression. He earned a computer science PhD at the University of Houston in 2007 and joined Google in 2008. He coauthored a 3D-aided face-recognition method, YouTube-8M, the AVA action dataset, recurrent and generative image-compression systems, and a visual-memory approach to image classification. He helped organize the THUMOS-2014 challenge and CLIC learned-compression workshops. As of October 2026, he is a researcher at Google DeepMind.

Works
64
Citations
16,941
h-index
33
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2007–2008

    A Houston PhD and Google

    He earns a computer science PhD from the University of Houston in 2007, where his research focuses on 2D-to-3D face recognition, and joins Google in 2008.

  2. 2010

    3D-aided face recognition

    In 2010, he coauthors a method for 3D-aided 2D face recognition that uses bidirectional relighting to handle large pose and illumination changes.

  3. 2014

    A million-video study

    In 2014, he coauthors a convolutional-network study of video classification on one million YouTube videos across 487 classes. He helps organize the THUMOS-2014 challenge and serves as Deep Learning area co-chair for ACM Multimedia in 2014.

  4. 2016

    YouTube-8M and recurrent codecs

    In 2016, he coauthors YouTube-8M, a video-classification benchmark of about eight million videos with 4,800 visual entities, and a recurrent-network framework for progressive, variable-rate image compression.

  5. 2017

    The AVA action dataset

    In 2017, he coauthors AVA, a dataset of 80 atomic visual actions across 430 fifteen-minute video clips, with 1.58 million labels localized in space and time.

  6. 2018–2021

    Learned image compression

    From 2018 to 2021, he helps organize the Workshop and Challenge on Learned Image Compression (CLIC) at CVPR. In 2020, he coauthors High-Fidelity Generative Image Compression, which combines generative adversarial networks with learned compression to create a generative lossy compression system.

  7. 2025

    Visual memory for image classification

    In 2025, he coauthors Towards flexible perception with visual memory, which breaks image classification into similarity from a pre-trained embedding and search through nearest-neighbor retrieval. Its visual memory supports adding data at different scales and removing it through unlearning and memory pruning.

  8. As of October 2026

    Research at Google DeepMind

    As of October 2026, he is a researcher at Google DeepMind.

Career

  1. NowGoogle DeepMindResearcher
  2. 2008–2023Google

Education

  1. 2007University of HoustonPhD in computer science

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

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes George Toderici. 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.
  2. 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.
  3. 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.
  4. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes Google. We estimate that 539 of the 6,766 people with a known current organization (8%) hold both a university role and a company role, including 20% of people at companies and 13% of people at universities.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of George Toderici yet. Send us a free-licensed photo.

  1. About Us archive.compression.cc. Checked 22 October 2026.
  2. 3D-aided 2D Face Recognition research.google. Checked 22 October 2026.
  3. Large-scale Video Classification with Convolutional Neural Networks research.google. Checked 22 October 2026.
  4. George Toderici research.google. Checked 22 October 2026.
  5. YouTube-8M: A Large-Scale Video Classification Benchmark arxiv.org. Checked 22 October 2026.
  6. Variable Rate Image Compression with Recurrent Neural Networks research.google. Checked 22 October 2026.
  7. AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions arxiv.org. Checked 22 October 2026.
  8. High-Fidelity Generative Image Compression arxiv.org. Checked 22 October 2026.
  9. Towards flexible perception with visual memory proceedings.mlr.press. Checked 22 October 2026.
  10. gtoderici (George Toderici) · GitHub github.com. Checked 22 October 2026.