PeopleofSuperintelligence

Researcher

Despoina Paschalidou

Despoina Paschalidou is a computer vision and graphics researcher whose work spans 3D reconstruction, scene synthesis and video generation. She earned a Diploma in Electrical and Computer Engineering at Aristotle University of Thessaloniki and completed a PhD at the Max Planck ETH Center for Learning Systems in December 2021 under Andreas Geiger and Luc van Gool. RayNet addressed volumetric reconstruction, ATISS generated indoor layouts, ReMatching integrated deformation priors into dynamic reconstruction, and Motive traced training clips affecting motion quality before receiving an ICML 2026 Outstanding Paper Honorable Mention. She began a full-time NVIDIA role in May 2024 and is a Senior Research Scientist at its Spatial Intelligence Lab; in 2026, she coauthored WorldTrace, a NeurIPS paper on memory for video world models.

Now at
NVIDIA
Badges

Timeline

  1. 2018

    RayNet for 3D reconstruction

    In 2018, she coauthors RayNet, a CVPR paper on volumetric 3D reconstruction with ray potentials.

  2. By 2021

    Electrical engineering and a PhD

    Before defending her PhD at the Max Planck ETH Center for Learning Systems in December 2021 under Andreas Geiger and Luc van Gool, she earns a Diploma in Electrical and Computer Engineering from Aristotle University of Thessaloniki.

  3. 2021

    ATISS for indoor layouts

    In 2021, she coauthors ATISS, an autoregressive transformer that generates diverse indoor environments from a room type and floor plan.

  4. February 2022

    Postdoctoral work at Stanford

    In February 2022, she starts a postdoctoral role at Stanford with Leonidas Guibas in the Geometric Computation Group. Her Stanford work concerns perceiving, manipulating and recreating 3D objects, humans and scenes.

  5. 2025

    ReMatching for dynamic reconstruction

    At ICLR 2025, she coauthors ReMatching, a framework that integrates deformation priors into dynamic reconstruction models to improve generalization to unseen views and timestamps.

  6. 2026

    Motive and motion attribution

    In 2026, she coauthors Motive, a data attribution framework that identifies training clips affecting motion quality in video generation. The paper receives an ICML 2026 Outstanding Paper Honorable Mention.

  7. 2026

    WorldTrace for video world models

    At NeurIPS 2026, she coauthors WorldTrace, a training-free memory framework that keeps compressed memory addressable in interactive video world models.

  8. Since May 2024

    Senior Research Scientist at NVIDIA

    She starts a full-time role as a Senior Research Scientist at NVIDIA in May 2024. As of October 2026, she works in the Spatial Intelligence Lab and focuses on computer vision and graphics.

Career

  1. Since 2024NVIDIASenior Research Scientist
  2. From 2022Stanford UniversityPostdoctoral Researcher

Education

  1. 2021Max Planck–ETH Center for Learning SystemsPhD
  2. Aristotle University of ThessalonikiDiploma in Electrical and Computer Engineering

From the biography.

  1. 16% of AI career moves in the 2020s went from a university to a companyIncludes Despoina Paschalidou. 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.
  2. Google has the highest combined h-index among AI organizationsIncludes NVIDIA. 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.
  3. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes NVIDIA. 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.
  4. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes NVIDIA. We estimate that alumni of 17 major AI labs went on to found or lead 350 other organizations, including 226 they founded, and we trace 472 career moves directly between 23 labs since 2010.

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 Despoina Paschalidou yet. Send us a free-licensed photo.

  1. Despoina Paschalidou paschalidoud.github.io. Checked 22 October 2026.
  2. RayNet: Learning Volumetric 3D Reconstruction With Ray Potentials openaccess.thecvf.com. Checked 22 October 2026.
  3. EECS Rising Stars 2023 | Welcome to Rising Stars eecsrisingstars2023.cc.gatech.edu. Checked 22 October 2026.
  4. ATISS: Autoregressive Transformers for Indoor Scene Synthesis papers.neurips.cc. Checked 22 October 2026.
  5. ReMatching Dynamic Reconstruction Flow research.nvidia.com. Checked 22 October 2026.
  6. Motion Attribution for Video Generation | Research research.nvidia.com. Checked 22 October 2026.
  7. Announcing the ICML 2026 Awards – ICML Blog blog.icml.cc. Checked 22 October 2026.
  8. WorldTrace: Addressable Memory for Video World Models research.nvidia.com. Checked 22 October 2026.