General Information

Name Anurag Ranjan
Summary Research Scientist at Apple working on Generative AI, 3D Geometry and Efficient Neural Nets.
Email anurag.ranjan(at)tue(dot)mpg(dot)de


  • 2016-2019
    Doctor of Philosophy
    Max Planck Institute for Intelligent Systems / University of Tübingen
    • Thesis
      • Towards geometric understanding of Motion. Summa cum laude (mit Auszeichnung, with honors, ∼top 5%)
  • 2013-2015
    Master of Science
    The University of British Columbia, Vancouver
    • Computer Science
  • 2009-2013
    Bachelor of Technology
    National Institute of Technology, Karnataka
    • Electronics and Communications Engineering


  • 2020 - Now
    Research Scientist
    • Led research efforts in Neural Radiance Fields, Generative AI and efficient neural networks.
    • Co-inventor of MobileOne and FastViT; fastest neural networks on iPhone, shipped across multiple products.
    • Co-inventor of FaceLit, Neuman; state-of-the-art generative models for faces and humans using NeRFs.
    • Co-developed Hypersim, most photorealistic and first large scale dataset released by Apple
  • Fall 2018
    Research Intern
    • Developed Competitive Collaboration, a general framework that facilitates competition and collaboration between different networks so that they learn from each other.
    • Demonstrated state-of-the-art results on joint unsupervised learning of depth, camera motion, optical flow and motion segmentation leading to a CVPR paper.
  • Summer 2017
    Research Intern
    • Developed a patented method for unsupervised video segmentation.
  • 2015
    Software Developer
    Mashup Machine

Positions Served

  • Area Chair
    • British Machine Vision Conference (BMVC) 2021, 2022, 2023.
    • 3D Vision (3DV) 2024.
    • Winter Conference on Applications of Computer Vision (WACV) 2024.
  • Reviewer

Honors and Awards

  • 2013-2015
    • Mitacs Globalink Graduate Fellowship
  • 2012
    • Mitacs Globalink Fellowship
    • Prime Minister's Letter of Appreciation