Leveraging Large-Scale Uncurated Data for Unsupervised Pre-training of Visual Features

Abstract : Pre-training general-purpose visual features with con-volutional neural networks without relying on annotations is a challenging and important task. Most recent efforts in unsupervised feature learning have focused on either small or highly curated datasets like ImageNet, whereas using uncurated raw datasets was found to decrease the feature quality when evaluated on a transfer task. Our goal is to bridge the performance gap between unsupervised methods trained on curated data, which are costly to obtain, and massive raw datasets that are easily available. To that effect , we propose a new unsupervised approach which leverages self-supervision and clustering to capture complementary statistics from large-scale data. We validate our approach on 96 million images from YFCC100M, achieving state-of-the-art results among unsupervised methods on standard benchmarks, which confirms the potential of unsu-pervised learning when only uncurated data are available. We also show that pre-training a supervised VGG-16 with our method achieves 74.6% top-1 accuracy on the validation set of ImageNet classification, which is an improvement of +0.7% over the same network trained from scratch.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02119564
Contributor : Mathilde Caron <>
Submitted on : Friday, May 3, 2019 - 7:30:52 PM
Last modification on : Tuesday, May 14, 2019 - 9:39:06 PM

File

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02119564, version 1

Collections

Citation

Mathilde Caron, Piotr Bojanowski, Julien Mairal, Armand Joulin. Leveraging Large-Scale Uncurated Data for Unsupervised Pre-training of Visual Features. 2019. ⟨hal-02119564⟩

Share

Metrics

Record views

43

Files downloads

101