PeopleofSuperintelligence

Researcher

Matthew Loper

Also known as Matt Loper, Matthew Maverick Loper

Matthew Loper is a computer vision researcher and the first author of SMPL, a learned model of human body shape and pose that won a SIGGRAPH Asia Test of Time Award in 2025; its team also received the 2025 PAMI Mark Everingham Prize. After earning an ScM at Brown University in 2008, he worked on parametric body modelling in the Perceiving Systems department of the Max Planck Institute for Intelligent Systems, coauthoring the OpenDR differentiable renderer and the MoSh motion-capture method in 2014. While working as a research engineer at Industrial Light and Magic in 2015, he published SMPL with Naureen Mahmood, Javier Romero, Gerard Pons-Moll and Michael Black, and he completed his doctorate at the University of Tübingen in 2017. He went on to coauthor FACSIMILE in 2019, which estimates a detailed body from a single photo, and, at Amazon, SPICE in 2021, which shows the person in one image in a new pose. In November 2025, he coauthored a framework that evaluates robot manipulation policies in Gaussian-splatting simulations of soft-body interactions.

Works
27
Citations
6,727
h-index
17
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2008

    An ScM from Brown

    He receives an ScM from Brown University in 2008 and turns to parametric body modelling, which he pursues as a PhD candidate in the Perceiving Systems department of the Max Planck Institute for Intelligent Systems.

  2. 2014

    OpenDR and MoSh

    He and Michael J. Black publish OpenDR at the European Conference on Computer Vision in 2014; the approximate differentiable renderer lets users express a forward graphics model and obtain derivatives with respect to its parameters. MoSh, with Naureen Mahmood and Black, appears in ACM Transactions on Graphics in 2014 and turns a motion-capture system into an approximate body scanner.

  3. 2015

    Research engineer at ILM

    In September 2015, he gives a VASC Seminar on OpenDR at Carnegie Mellon's Robotics Institute while working as a research engineer at Industrial Light and Magic. He is completing his PhD in the Perceiving Systems department of the Max Planck Institute for Intelligent Systems, with research on differentiable rendering and statistical body shape modelling.

  4. 2015

    The SMPL body model

    At SIGGRAPH Asia 2015, he and Naureen Mahmood, Javier Romero, Gerard Pons-Moll and Michael J. Black present SMPL, the Skinned Multi-Person Linear model. Learned from data, the skinned vertex-based model represents a wide variety of body shapes in natural human poses and works with existing graphics pipelines.

  5. 2017

    A doctorate from Tübingen

    He earns a Dr. rer. nat. from the University of Tübingen in 2017 with the dissertation Human Shape Estimation Using Statistical Body Models, with Michael J. Black as reviewer.

  6. 2019–2021

    FACSIMILE and SPICE

    In 2019, he coauthors FACSIMILE, which estimates a detailed body from a single photo using a loss on per-pixel surface normals. At Amazon, he coauthors SPICE for ICCV 2021, a self-supervised method that uses 3D body information to generate an image of a person in a new target pose.

  7. 2025

    Two awards for SMPL

    In October 2025, SMPL receives the PAMI Mark Everingham Prize at ICCV 2025, an honour for contributions of significant benefit to the computer vision community that goes to Pons-Moll and his coauthors Black, Loper, Mahmood and Romero. SIGGRAPH Asia 2025 also gives the SMPL paper a Test of Time Award.

  8. November 2025

    Real-to-sim robot policy evaluation

    In November 2025, he coauthors a real-to-sim evaluation framework with researchers from Columbia University, SceniX Inc. and Google DeepMind. It builds soft-body digital twins from real-world videos and renders robots and scenes with 3D Gaussian Splatting to test manipulation policies on tasks such as plush toy packing and rope routing.

Career

  1. 2015Industrial Light & MagicResearch engineer

Education

  1. 2017University of TübingenDr. rer. nat.
  2. 2008Brown UniversityScM
  3. From 2008Max Planck Institute for Intelligent SystemsPhD candidate

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

Most-cited works

  1. SMPL2015 · ACM Transactions on Graphics · 3,728 citationsWith Gerard Pons-Moll and Naureen Mahmood
  2. SMPL: A Skinned Multi-Person Linear Model2023 · ACM eBooks · 770 citationsWith Gerard Pons-Moll and Naureen Mahmood
  3. FAUST: Dataset and Evaluation for 3D Mesh Registration2014 · IEEE Computer Society Conference on Computer Vision and Pattern Recognition · 690 citations
  4. OpenDR: An Approximate Differentiable Renderer2014 · Lecture notes in computer science · 457 citations
  5. MoSh2014 · ACM Transactions on Graphics · 344 citationsWith Naureen Mahmood

From OpenAlex, as of 21 October 2026.

  1. Google has the highest combined h-index among AI organizationsIncludes Max Planck Society. 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.
  2. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Max Planck Society. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  3. People shaping AI work at a median of 3 organizations over a career: 3 for university starters and institute and lab starters, versus 2 for company startersIncludes Max Planck Society. Across 6,853 people shaping AI with a dated work role, we estimate a median of 3 parent organizations over a career (mean 3.6), comparing role groups, starting sectors and starting decades from the 1950s to 2020–2026.
  4. The United States employs an estimated 48% of people shaping AIIncludes Max Planck Society. 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 Matthew Loper yet. Send us a free-licensed photo.

  1. OpenDR: An Approximate Differentiable Renderer - Robotics Institute Carnegie Mellon University ri.cmu.edu. Checked 20 October 2026.
  2. An Approximate Differentiable Renderer - Microsoft Research microsoft.com. Checked 20 October 2026.
  3. OpenDR: An Approximate Differentiable Renderer | ML Anthology mlanthology.org. Checked 20 October 2026.
  4. MoSh Dataset mosh.is.tue.mpg.de. Checked 20 October 2026.
  5. Modelling Human Bodies; Speakers · December 17-18th, 2015 - Santiago, Chile bodymodelling2015.is.tuebingen.mpg.de. Checked 20 October 2026.
  6. “SMPL: a skinned multi-person linear model” by Loper, Mahmood, Romero, Pons-Moll and Black – ACM SIGGRAPH HISTORY ARCHIVES history.siggraph.org. Checked 20 October 2026.
  7. Matthew Loper - The Mathematics Genealogy Project mathgenealogy.org. Checked 20 October 2026.
  8. ICCV 2019 Open Access Repository: FACSIMILE: Fast and Accurate Scans From an Image in Less Than a Second openaccess.thecvf.com. Checked 20 October 2026.
  9. ICCV 2021 Open Access Repository: Learning Realistic Human Reposing Using Cyclic Self-Supervision With 3D Shape, Pose, and Appearance Consistency openaccess.thecvf.com. Checked 20 October 2026.
  10. [2110.05458] Learning Realistic Human Reposing using Cyclic Self-Supervision with 3D Shape, Pose, and Appearance Consistency ar5iv.labs.arxiv.org. Checked 20 October 2026.
  11. Real Virtual Humans | Real Virtual Humans | University of Tübingen virtualhumans.mpi-inf.mpg.de. Checked 20 October 2026.
  12. Award Winners | SIGGRAPH Asia 2025 asia.siggraph.org. Checked 20 October 2026.
  13. Real-to-Sim Robot Policy Evaluation with Gaussian Splatting Simulation of Soft-Body Interactions arxiv.org. Checked 20 October 2026.