Postdoc
Title: Postdoc in 3D Computer Vision / Machine Learning
Location: Ecole Polytechnique, France
Start date: December 2025 - February 2026
Description:
A postdoctoral researcher position is available
at the LIX research laboratory of Ecole Polytechnique in France, in
the area of:
"Foundation Models for Geometric Data"
The goal of this project is to develop tools (novel architectures, training strategies, data curration etc.) for
creating foundation models capable of zero shot and few shot performance on a broad range of tasks involving 3D data.
While foundation models are readily available for images and texts, their relevance for geometric (3D) data remains
limited, despite potentially significant applicability.
The ideal candidates should have strong background in at least some of
the following:
- Computer Vision, Computer Graphics, 3D Shape Analysis
- Machine Learning
- Generative Modeling, especially for 3D data
- Theoretical aspects of shape analysis
Candidates must have a PhD at the starting time of the postdoc, and a strong publication record in top venues in Machine Learning
(NeurIPS, CVPR, ICCV, etc.). The expected duration of the postdoc is 2 years.
Venue and supervision:
Successful candidates will be based at Ecole Polytechnique located in
Palaiseau, Paris area, France. The project will be supervised by
Maks
Ovsjanikov (LIX, Ecole Polytechnique).
How to apply:
To apply, please contact Maks Ovsjanikov
(maks@lix.polytechnique.fr) and include
[Computer Vision postdoc application]
in the subject line.
In your email, please include: 1) your CV
with a publication list and 2) a short cover letter. In your cover letter,
please
make sure to explicitly describe your experience in the 4 topics mentioned above.
Reference letters are optional but may need to be provided upon request.
PhD Students
Note: I am always looking for excellent PhD candidates with strong background in
mathematics and computer science. In virtually all cases, prospective PhD student start by
doing a 5-6 month M2 internship in our group.
If you are interested in applying for a PhD
position, the best way to apply is via the ELLIS PhD
program that I am a part of.
Alternatively, if you are enrolled in an M2 program (e.g.,
MVA), please feel free to reach out if you are interested in doing a pre-PhD internship in our
group. This typically happens at the end of the Master's program, as the final research internship.
Lastly, if you are simply passionate about the work done in our group, you can always reach
out to me directly. The last option is the least commonly successful, however.
Internships
Note 1: Unfortunately, I do not accept undergraduate internship requests. My only internship offers are for
students currently enrolled in, or having completed their Master's degrees. I am sorry for not being able
to respond to each individual undergraduate internship request.
Note 2: I participate in
the
Research
Program for International Talents for Master's-level candidates. You can apply either
through that program, or by writing to me directly.
As mentioned above, I typically provide
pre-PhD internships to students enrolled in the final year of their Master's (either in France or abroad).
If you are
interested in such an internship, please feel free to reach out to me directly.
Position: Research internship (Master's or PhD student)
Title: Novel Deep Learning approaches for 3D shape synthesis and analysis
Location: Ecole Polytechnique, France
Start date: March — April 2026
Description:
Machine Learning methods have revolutionized image and text processing.
Compared to this success, progress in 3D shape understanding and generation lags
behind. Nevertheless, several recent approaches have been proposed for creating realistic 3D
models using Denoising Diffusion techniques in particular, and shape analysis using Geometric
Deep Learning.
The goal of this research project will be to develop new algorithms for accurate 3D shape
synthesis andanalysis, by building upon recent advances in both standard text and 2D Computer
Vision (so-called foundation models) and recent progress in 3D shape understanding. Ultimately,
the goal will be to pushthe state of the art in terms of accuracy and realism of produced
models, and 3D shape understanding techniques.
The ideal candidates should have strong background in
at least some of the following:
- Computer Vision or Computer Graphics
- Machine Learning
- Differential Geometry
- Geometry Processing, and 3D shape analysis (point clouds, meshes, volumetric grids, etc.)
Note: We prioritize students who combine a strong background in both mathematics and computer science, especially with prior research experience.
The ultimate goal of this internship is to transition to a PhD position in our group. The PhD has funding guaranteed by the ERC Consolidator Grant VEGA
Successful candidates will be based at Ecole Polytechnique located in Palaiseau,
Paris area, France. The project will be supervised by Maks Ovsjanikov.
How to apply:
To apply, please contact Maks Ovsjanikov: maks@lix.polytechnique.fr, and put
[internship application] in the subject line. In your email, please include your CV,
the transcript of courses taken with the grades and a short cover letter. In your
cover letter, please describe how your experience relates to the topics in this
announcement.