Julian Ost - Submission to the CVPR 2026 Doctoral Consortium

Julian Ost

PhD University: Princeton University

Graduation Date: November 2026

PhD Advisor: Felix Heide

Google Scholar: https://scholar.google.com/citations?user=mUbWwU4AAAAJ

Poster Title: 3D Scene Generation and Understanding Through Generative Priors

First Author Papers: 2

Requested Mentors Instructions: I understand — my preferred mentors are listed below.

Requested Mentor 1: Christian Rupprecht, [e-mail]

Requested Mentor 2: Robin Rombach, [e-mail]

Requested Mentor 3: Vladlen Koltun, [e-mail]

Requested Mentor 4: Alexandre Alahi, [e-mail]

Requested Mentor 5: Andrea Vedaldi, [e-mail]

Requested Mentor 6: Christoph Lassner, [e-mail]

Requested Mentor 7: Andreas Geiger, [e-mail]

Requested Mentor 8: Matthias Nießner, [e-mail]

Academia Or Industry: Undecided

Mentor Questions:

- Given the recent discourse on video models, latent world models and explicit 3D understanding and models. Which role do you think will each of these models take in the foreseeable future?

- The gap in compute and data resources between a academic lab, a seed-stage startup, and big industry labs is enormous. Working on generative models and world models are also becoming more and more resource-hungry. On what kind of research and work would you realistically focus in any of the three settings and where do you think those settings should differ?

- If you were advising a researcher or company on where to invest in the next five years within 3D generation/understanding and world models for robotics and driving, what gap would you tell them is mostly ignored right now but will be of high importance?

CVPR Participation: I will be presenting my work at a workshop or tutorial, I will be attending as a general participant

Previous DC Certification: I certify that I have not previously attended a Doctoral Consortium at ICCV, ECCV or CVPR.

In Person Confirmation: I confirm that I plan to attend CVPR 2026 in person.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 107