News

• 08/2025 We release DINOv3 , a state-of-the-art computer vision model trained with self-supervised learning (SSL) that produces powerful, high-resolution image features. For the first time, a single frozen vision backbone outperforms specialized solutions on multiple long-standing dense prediction tasks.
• 05/2024 We release our paper on Automatic data curation for self-supervised learning , a generic, principled approach to build large, diverse and balanced training datasets. VentureBeat wrote an article and Andrew Ng's letter gave a detailed account about it.
• 04/2023 We release DINOv2 , a family of foundation models producing universal features suitable for image-level visual tasks (image classification, instance retrieval, video understanding) as well as pixel-level visual tasks (depth estimation, semantic segmentation).
• 11/2022 I am joining FAIR Labs, Meta as an AI Research Scientist.
Older news >>>
11/2022 I successully defended my thesis on "Annotation-efficient learning for object discovery and detection". The thesis was reviewed by Prof. Andrew Zisserman and Prof. Tinne Tuytelaars, examinated by Dr. Cordelia Schmid, Dr. Yannis Avrithis, Dr. Elena Sizikova (invited member) and Dr. Oriane Siméoni (invited member), and supervised by Prof. Jean Ponce and Dr. Patrick Pérez.
07/2022 Our work on active and weakly-supervised object detection, BiB, is accepted to ECCV 2022.
10/2021 Our work on unsupervised object discovery/detection, LOST, is accepted to BMVC 2021.
09/2021 Our work on large-scale unsupervised object discovery, LOD, is accepted to NeuRIPS 2021.
03/2021 Our paper on weakly supervised lesion segmentation is accepted at MIDL 2021.
07/2020 Our work on unsupervised object discovery, rOSD, is accepted at ECCV 2020.
10/2019 I am visiting Center for Data Science, New York University for 3 months.
03/2019 Our work on unsupervised object discovery, OSD, is accepted at CVPR 2019.
10/2018 I start my PhD in INRIA's Willow team and Valeo.ai under the supervision of Prof. Jean Ponce and Dr. Patrick Pérez.
04/2018 I start a 6-month research internship in the WILLOW team of INRIA and Center for Data Science (New York university), working on object discovery under the supervision of Prof. Jean Ponce and Prof. Yann LeCun.
04/2017 I start a 5-month internship at Technicolor, working on Image inpainting under the supervision of Dr. Patrick Pérez.

Invited Talks

• 10/2024 Automatic data curation, Vanderbilt Machine Learning Seminars.
• 06/2023 Annotation-efficient learning for object discovery and detection, VinAI seminar.
• 12/2021 Unsupervised Object Discovery, IMAGINE team, École des ponts ParisTech, Paris.

Publications