PeopleofSuperintelligence

Researcher

Jakob Engel

Also known as Jakob Julian Engel

Jakob Engel is a computer vision researcher whose direct methods for visual SLAM track a moving camera and map its surroundings straight from image intensities. During his PhD in Daniel Cremers's Computer Vision Group at the Technical University of Munich, he created LSD-SLAM, published at ECCV in 2014, and Direct Sparse Odometry, which appeared in 2016. LSD-SLAM received the ECCV 2024 Koenderink Award for standing the test of time. His 2023 paper as first author introduced Project Aria, a device for egocentric, multimodal AI research, and in 2024 he coauthored SceneScript, which reconstructs scenes as sequences of structured language commands. He is a Director of Research at Meta Reality Labs, where he leads egocentric machine perception and Spatial AI research for Project Aria.

Timeline

  1. 2009–2012

    A self-flying quadrocopter

    He earns a bachelor's degree in computer science at the Technical University of Munich (TUM) in 2009 and a master's degree in December 2011. Working with Juergen Sturm and Daniel Cremers, he writes a master's thesis that lets a low-cost Parrot AR.Drone quadrocopter fly autonomously in unknown environments using PTAM. In December 2012, the thesis wins the Siemens Award for the best master's thesis of 2012.

  2. 2012–2014

    LSD-SLAM at TUM

    In September 2012, he begins full-time PhD work in Daniel Cremers's Computer Vision Group at TUM. With Thomas Schöps and Cremers, he publishes LSD-SLAM at ECCV in 2014, a direct monocular SLAM method that works on image intensities for both tracking and mapping, estimates semi-dense depth maps and builds large-scale maps with loop closures.

  3. 2014

    EMVA Young Professional Award

    He receives the 2014 Young Professional Award of the European Machine Vision Association for his work on semi-dense visual odometry for a monocular camera. A 2014 paper by Thomas Schöps, Engel and Cremers on semi-dense visual odometry for augmented reality on a smartphone wins the Best Short Paper Award at ISMAR.

  4. 2016–2017

    Direct Sparse Odometry and a PhD

    He receives his PhD in computer science from TUM's Computer Vision Group in 2016. His Direct Sparse Odometry (DSO), developed partly during an internship with Vladlen Koltun at Intel and partly with Cremers, appears on arXiv in July 2016 and in Transactions on Pattern Analysis and Machine Intelligence in 2017. DSO jointly optimizes camera motion and geometry, integrates a full photometric calibration, and exceeds LSD-SLAM in runtime, accuracy and robustness.

  5. 2023–2024

    Project Aria and SceneScript

    In August 2023, he is first author of the paper introducing Project Aria, an egocentric, multimodal data recording and streaming device built by a team at Meta Reality Labs Research. In March 2024, he coauthors SceneScript, which produces full scene models as sequences of structured language commands with an autoregressive, token-based method trained on Aria Synthetic Environments, a synthetic dataset of 100,000 indoor scenes.

  6. 2024

    The Koenderink Award

    LSD-SLAM, written with Thomas Schöps and Daniel Cremers, receives the ECCV 2024 Koenderink Award for standing the test of time. At ECCV 2024, he gives a keynote at the ICVSE workshop, representing Meta Reality Labs in Seattle.

  7. October 2025

    Aria Gen 2 Pilot Dataset

    In October 2025, he coauthors the Aria Gen 2 Pilot Dataset, an open egocentric multimodal dataset captured with Aria Gen 2 glasses. Its first release covers cleaning, cooking, eating, playing and outdoor walking, with raw sensor data and outputs from machine perception algorithms.

  8. As of October 2026

    Director of Research at Meta

    He is a Director of Research at Meta Reality Labs, where he leads egocentric machine perception and Spatial AI research as part of Project Aria. His work includes shipping core localization technology into Meta's Quest and Orion product lines, and his interests include how AI models can be grounded in the physical 3D world.

  1. Google had the most people shaping AI every year from 2014 to 2025Our estimates of which organizations employed the most people shaping AI from 1961 to 2025, with a year-by-year chart.
  2. The United States employs an estimated 53% of people in our data with a known employment countryOur estimates of where AI careers are concentrated now and how countries’ shares of dated positions changed over time.
  3. 46% of living people in our data with known birth and employment countries work outside their birth countryOur estimate of cross-border careers in AI, from country of birth and the country of each current organization.
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionOur estimates of where the people shaping AI studied and where their careers led, from education records in cited biographies.

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 Jakob Engel yet. Send us a free-licensed photo.

  1. Jakob Engel | Computer Vision, Machine Learning, Robotics jakobengel.github.io. Checked 19 October 2026.
  2. Jakob Engel videolectures.net. Checked 19 October 2026.
  3. Computer Vision Group - Awards cvg.cit.tum.de. Checked 19 October 2026.
  4. Computer Vision Group - Visual SLAM - LSD-SLAM: Large-Scale Direct Monocular SLAM cvg.cit.tum.de. Checked 19 October 2026.
  5. [1607.02565] Direct Sparse Odometry arxiv.org. Checked 19 October 2026.
  6. [2308.13561] Project Aria: A New Tool for Egocentric Multi-Modal AI Research arxiv.org. Checked 19 October 2026.
  7. [2403.13064] SceneScript: Reconstructing Scenes With An Autoregressive Structured Language Model arxiv.org. Checked 19 October 2026.
  8. ICVSE @ ECCV 2024 - PROGRAM sites.google.com. Checked 19 October 2026.
  9. [2510.16134] Aria Gen 2 Pilot Dataset arxiv.org. Checked 19 October 2026.