0
Emerging Properties in Self-Supervised Vision Transformers
Mathilde Caron, Hugo Touvron, Ishan Misra et al., 2021. Proceedings of ICCV 2021.
Introduces DINO, a self-distillation method for training vision transformers without labels, and shows that the resulting features capture object layout and support nearest-neighbor retrieval.
What it's used for here
The inductive-methods survey proposes its self-supervised image features for clustering glyph images into an independent table of letter forms.
Cited by on this site 1 record
1 methods surveys
SRCInductive methods
Cite it
Caron, M., Touvron, H., Misra, I., Jégou, H., Mairal, J., Bojanowski, P., & Joulin, A. (2021). Emerging Properties in Self-Supervised Vision Transformers. Proceedings of ICCV 2021. https://arxiv.org/abs/2104.14294
@article{caron2021emerging,
author = {Caron, Mathilde and Touvron, Hugo and Misra, Ishan and Jégou, Hervé and Mairal, Julien and Bojanowski, Piotr and Joulin, Armand},
title = {Emerging Properties in Self-Supervised Vision Transformers},
journal = {Proceedings of ICCV 2021},
year = {2021},
url = {https://arxiv.org/abs/2104.14294}
}TY - JOUR TI - Emerging Properties in Self-Supervised Vision Transformers AU - Caron, Mathilde AU - Touvron, Hugo AU - Misra, Ishan AU - Jégou, Hervé AU - Mairal, Julien AU - Bojanowski, Piotr AU - Joulin, Armand PY - 2021 JO - Proceedings of ICCV 2021 UR - https://arxiv.org/abs/2104.14294 ER -
Caron, Mathilde et al.. “Emerging Properties in Self-Supervised Vision Transformers.” Proceedings of ICCV 2021 (2021).