Featurising Pixels from Dynamic 3D Scenes with Linear In-Context Learners


Nikita Araslanov1,2,3
Martin Sundermeyer1
Hidenobu Matsuki1
David Joseph Tan1
Federico Tombari1,2,3
1 Google       2 TU Munich       3 Munich Center for Machine Learning

✨ CVPR 2026 Oral Presentation ✨


Paper Supplemental Material (26MB) Code
The gist: Learning dense representations from videos with linear in-context learning.

Image frames
LILA (PCA)

Citation

@inproceedings{Araslanov:2025:LILA,
  author = {Araslanov, Nikita and Sundermeyer, Martin and Matsuki, Hidenobu and Tan, David Joseph and Tombari, Federico},
  title = {Featurising Pixels from Dynamic 3D Scenes with Linear In-Context Learners},
  booktitle = {CVPR},
  year = {2026},
}