Semantic Scholar Open Access 2023 7509 sitasi

DINOv2: Learning Robust Visual Features without Supervision

M. Oquab Timothée Darcet Théo Moutakanni Huy V. Vo Marc Szafraniec +21 lainnya

Abstrak

The recent breakthroughs in natural language processing for model pretraining on large quantities of data have opened the way for similar foundation models in computer vision. These models could greatly simplify the use of images in any system by producing all-purpose visual features, i.e., features that work across image distributions and tasks without finetuning. This work shows that existing pretraining methods, especially self-supervised methods, can produce such features if trained on enough curated data from diverse sources. We revisit existing approaches and combine different techniques to scale our pretraining in terms of data and model size. Most of the technical contributions aim at accelerating and stabilizing the training at scale. In terms of data, we propose an automatic pipeline to build a dedicated, diverse, and curated image dataset instead of uncurated data, as typically done in the self-supervised literature. In terms of models, we train a ViT model (Dosovitskiy et al., 2020) with 1B parameters and distill it into a series of smaller models that surpass the best available all-purpose features, OpenCLIP (Ilharco et al., 2021) on most of the benchmarks at image and pixel levels.

Topik & Kata Kunci

Penulis (26)

M

M. Oquab

T

Timothée Darcet

T

Théo Moutakanni

H

Huy V. Vo

M

Marc Szafraniec

V

Vasil Khalidov

P

Pierre Fernandez

D

Daniel Haziza

F

Francisco Massa

A

Alaaeldin El-Nouby

M

Mahmoud Assran

N

Nicolas Ballas

W

Wojciech Galuba

R

Russ Howes

P

Po-Yao (Bernie) Huang

S

Shang-Wen Li

I

Ishan Misra

M

Michael G. Rabbat

V

Vasu Sharma

G

Gabriel Synnaeve

H

Hu Xu

H

H. Jégou

J

J. Mairal

P

Patrick Labatut

A

Armand Joulin

P

Piotr Bojanowski

Format Sitasi

Oquab, M., Darcet, T., Moutakanni, T., Vo, H.V., Szafraniec, M., Khalidov, V. et al. (2023). DINOv2: Learning Robust Visual Features without Supervision. https://doi.org/10.48550/arXiv.2304.07193

Akses Cepat

Lihat di Sumber doi.org/10.48550/arXiv.2304.07193
Informasi Jurnal
Tahun Terbit
2023
Bahasa
en
Total Sitasi
7509×
Sumber Database
Semantic Scholar
DOI
10.48550/arXiv.2304.07193
Akses
Open Access ✓