Diane Larlus
Computer Vision & Machine Learning Research Scientist
About Me
Principal Research Scientist and Team Lead at NAVER LASB Europe.
Research on lifelong learning for visual representations, with applications
to the semantic and geometrical understanding of complex scenes.
Recently worked on model distillation and egocentric video understanding.
Research on lifelong learning for visual representations, with applications
to the semantic and geometrical understanding of complex scenes.
Recently worked on model distillation and egocentric video understanding.
Recent projects
Geo4D @ICCV25
Geo4D repurposes a video diffusion model for monocular 4D reconstruction. Collaboration with VGG, Oxford. ICCV 2025
LUDVIG @ICCV25
Uplifts DINOv2, SAM and CLIP in 3D with a simple learning-free rendering process. Collaboration with Thoth, Inria. ICCV 2025
DUNE @CVPR25
A single encoder distilled from multiple teachers: DINOv2, MASt3R and Multi-HMR, versatile enough to perform heterogeneous tasks. CVPR 2025
UNIC @ECCV24
A UNIC classification model that distills from strong pretrained models, and performs on par of better then each of them. ECCV 2024
PANDAS @CoLLAs24
Automatically discover novel classes from unlabeled images and add the discovered classes to extend an existing detector. CoLLAs 2024
EPIC Fields @NeurIPS23
Extension of EPIC KITCHENS dataset with 3D information. Collaboration with VGG, Oxford and MaVi, Bristol. NeurIPS 2023
More projects:
Here is a list of all projects
News
2025
- 2 papers accepted at ICCV5
- LUDVIG: Learning-Free Uplifting of 2D Visual Features to Gaussian Splatting Scenes [Project page, code]
- Geo4D: Leveraging Video Generators for Geometric 4D Scene Reconstruction [Project page, code]
- Invited as a Keynote at the CVPR in Paris local meeting prior to CVPR2025
- 2 papers accepted at the CVPR25
- DUNE: Distilling a Universal Encoder from Heterogenous 2D and 3D Teachers [Project page]
- Layered motion fusion: Lifting motion segmentation to 3D in egocentric videos
- Invited as a lecturer for the 2nd ELLIS Winter School on Foundation Models (FoMo25)
- Keynote speaker at the VISIGRAPP 2025 conference
- Area Chair for ICCV25
2024
- Invited speaker at two ECCV 2024 workshops: the Computer Vision for Metaverse and the Transparent & Reflective objects In the wild (TRICKY) ones
- Paper accepted at ECCV 2024: UNIC: Universal Classification Models via Multi-teacher Distillation [Project page]
- Paper accepted at CoLLAs 2024: PANDAS: Prototype-based Novel Class Discovery and Detection [Paper, code]
- Invited for a keynote at the CoLLAs 2024 conference
- Paper accepted at TMLR: On Good Practices for Task-Specific Distillation of Large Pretrained Visual Models [Paper]
- Invited as a lecturer for the International Computer Vision Summer School ICVSS 2024
- Invited speaker at the First Joint Egocentric Vision (EgoVis) workshop at CVPR24
- Paper accepted at ICLR24: Weatherproofing Retrieval for Localization with Generative AI & Geometric Consistency [Paper, project page]
- Area Chair for CVPR24, ECCV24
2023
- Invited speaker at the Self-Supervised Learning - Theory and Practice at NeurIPS23
- Co-organizing the 3rd NAVER LABS Europe International Workshop on AI for Robotics
- EPIC Fields accepted at NeurIPS23 benchmark and dataset track [project page]
- Invited speaker at the workshop on Assistive Computer Vision and Robotics at ICCV23
- Release of the EPIC Fields dataset and benchmark [project page]
- Paper accepted at CoLLAs23: RaSP: Relation-aware Semantic Prior for Weakly Supervised Incremental Segmentation [paper]
- 2 papers accepted at CVPR23
- Fake it till you make it: learning transferable representations from synthetic ImageNet clones [project page]
- SLACK: Stable Learning of Augmentations with Cold-start and KL regularization [project page]
- Keynote speaker at two CVPR23 workshops, the Continual Learning in Computer vision and the Earth Vision ones
- Paper accepted at ICLR23: No Reason for No Supervision: Improved Generalization in Supervised Models [paper, project page]
- Doctoral Consortium chair for ICCV23
- Area Chair for ICCV23
More
- see Older news
- see Full list of publications
Recent talks
| Jun 2025 | Keynote Speaker at CVPR in Paris - Breaking the mold: Teaching Visual Models to Generalize | |
| Jun 2025 | Invited talk at the ENPC Imagine team retreat - Breaking the mold: Teaching Visual Models to Generalize | |
| May 2025 | Invited talk at the Thoth team at Inria - Learning versatile visual representations | |
| Mar 2025 | Lecture at the ELLIS Winter School on Foundation Models (FoMo) - Lifelong learning in the Age of Foundation Models | |
| Fev 2025 | Keynote Speaker at the Joint VISIGRAPP Conference - Lifelong Visual Representation Learning | |
| Oct 2024 | Invited talk at the Transparent & Reflective objects In the wild (TRICKY) workshop at ECCV24 - Towards a Geometric Understanding in Egocentric Videos | |
| Oct 2024 | Invited talk at the Computer Vision for Metaverse workshop at ECCV24 - Towards a Geometric Understanding in Egocentric Videos | |
| Aug 2024 | Keynote Speaker at the Third Conference on Lifelong Learning Agents (CoLLAs) 2024 - Lifelong learning for visual representations [Video] | |
| Jul 2024 | Lecture at the International Computer Vision Summer School (ICVSS) 2024 - Lifelong learning for visual representations | |
| Jun 2024 | Invited talk at the First Joint Egocentric Vision (EgoVis) at CVPR24 - Towards a Geometric Understanding in Egocentric Videos [Video] | |
| Dec 2023 | Invited talk at the Self-Supervised Learning - Theory and Practice workshop at NeurIPS23 - Learning visual representations that transfers [Video] | |
| Oct 2023 | Invited talk at the Assistive Computer Vision and Robotics workshop at ICCV23 - Towards a Geometric Understanding in Egocentric Videos | |
| Jun 2023 | Invited talk at the Continual Learning in Computer Vision workshop at CVPR23 - Lifelong visual representation learning [Video] | |
| Jun 2023 | Invited talk at the EarthVision workshop at CVPR23 - Lifelong visual representation learning | |
| Sep 2022 | Invited talk at the Thoth team at Inria - Towards Semantic Understanding in Egocentric Videos | |
| Sep 2022 | Invited talk at the Linkmedia Speaks Science seminar series - Lifelong visual representation learning [Video] | |
| Jul 2022 | Invited talk at the 1st International Workshop on Pervasive Artificial Intelligence - Lifelong visual representation learning | |
| Jul 2022 | Keynote Speaker at CAp-RFIAP 2022 - Lifelong visual representation learning | |
| Jun 2022 | Invited talk at the O-DRUM Workshop at CVPR22 - Using Text in Computer Vision | |
| Apr 2022 | Invited talk at GDR TAL Day - Learning transferable visual representations | |
| Jan 2022 | Invited talk at KUIS AI Meetings - Towards Semantic Understanding in Egocentric Videos | |
| Jun 2021 | Invited talk at the Women in Computer Vision (WiCV) Workshop at CVPR21 - Lifelong visual representation learning [Video] | |
| May 2021 | Invited talk at the ANITI research institute (scientific seminar) - Learning from weak supervision and mitigating catastrophic forgetting | |
| Oct 2020 | Keynote Speaker for MIAI Days - Lifelong visual representation learning | |
| Aug 2020 | Invited talk at the ILR Workshop at ECCV20 - From Instance-Level to Semantic Image Retrieval [Video] | |
| Nov 2019 | Invited talk at the SOLARIS Workshop - Self-supervised learning for category-level geometry estimation | |
| Apr 2019 | Invited talk at the StatLearn 19 conference - Learning Image Representations for Efficient Visual Search | |
| Nov 2018 | Invited speaker, AI and Vision session at SophIA Summit - Visual Search | |
| Jul 2018 | Lecture at the PRAIRIE Artificial Intelligence summer school (PAISS 2018) - Visual Search | |
| Older: see CV |
Teaching
| 2025/2026 | Perception, Vision et Apprentissage (Partie apprentissage continu de représentations visuelles) - master level - ENSIMAG, Grenoble | |
| Mar 2025 | Speaker at the ELLIS Winter School on Foundation Models, Amsterdam | |
| 2024/2025 | Perception, Vision et Apprentissage (Partie apprentissage continu de représentations visuelles) - master level - ENSIMAG, Grenoble | |
| Aug 2024 | Lecture on Continual Learning at the International Computer Vision Summer School, Sicily (ICVSS 2024) | |
| 2024 | Découvrir l’IA par le jeu (Discover AI with games, a 2 day training session for high school teachers) - Maison pour la Science Alpes-Dauphiné | |
| 2023/2024 | L’apprentissage continu de représentations visuelles - master level - ENSIMAG, Grenoble [webpage] | |
| Dec 2023 | Lifelong visual representation learning - Master AI 4 One Health, UGA, Grenoble | |
| Sep 2023 | Keynote speaker at the VISMAC International Summer School on Machine Vision (VISMAC 2023) | |
| 2023 | Découvrir l’IA par le jeu (Discover AI with games, a 2 day training session for high school teachers) - Maison pour la Science Alpes-Dauphiné | |
| 2022/2023 | L’apprentissage continu de représentations visuelles - master level - ENSIMAG, Grenoble | |
| Dec 2022 | Lifelong visual representation learning - Master AI 4 One Health, UGA, Grenoble | |
| 2022 | Découvrir l’IA par le jeu (Discover AI with games, a 2 day training session for high school teachers) - Maison pour la Science Alpes-Dauphiné | |
| Jan 2022 | Guest lecture on Visual Search - Posts and Telecommunications Institute of Technology (PTIT), Vietnam | |
| 2021/2022 | Comprendre des données visuelles à grande échelle - master level - ENSIMAG, Grenoble | |
| Nov 2021 | Lifelong visual representation learning - Master AI 4 One Health, UGA, Grenoble | |
| May 2021 | Guest lecture on Visual Search - Hanoi University of Science and Technology (HUST), Vietnam | |
| 2020/2021 | Comprendre des données visuelles à grande échelle - ENSIMAG, Grenoble | |
| 2020 | Découvrir l’IA par le jeu (Discover AI with games, a 2 day training session for high school teachers) - Maison pour la Science Alpes-Dauphiné | |
| Nov 2019 | Lecture on Visual Search as part of the ‘Deep Learning’ course for the MVA master - ENS Paris | |
| 2019/2020 | Comprendre des données visuelles à grande échelle - master level - ENSIMAG, Grenoble | |
| Feb 2019 | Lecture on Visual Search as part of the ‘Deep Learning in Practice’ course for the MVA master - ENS Paris | |
| 2018/2019 | Comprendre des données visuelles à grande échelle - master level - ENSIMAG, Grenoble | |
| July 2018 | Lecture on Visual Search at the PRAIRIE Artificial Intelligence summer school (PAISS 2018) | |
| Older: see CV |