Corpus ID: 211532737

Learning Representations by Predicting Bags of Visual Words

@article{Gidaris2020LearningRB,
  title={Learning Representations by Predicting Bags of Visual Words},
  author={Spyros Gidaris and Andrei Bursuc and Nikos Komodakis and Patrick P{\'e}rez and Matthieu Cord},
  journal={ArXiv},
  year={2020},
  volume={abs/2002.12247}
}
  • Spyros Gidaris, Andrei Bursuc, +2 authors Matthieu Cord
  • Published 2020
  • Computer Science
  • ArXiv
  • Self-supervised representation learning targets to learn convnet-based image representations from unlabeled data. Inspired by the success of NLP methods in this area, in this work we propose a self-supervised approach based on spatially dense image descriptions that encode discrete visual concepts, here called visual words. To build such discrete representations, we quantize the feature maps of a first pre-trained self-supervised convnet, over a k-means based vocabulary. Then, as a self… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 78 REFERENCES

    Unsupervised Visual Representation Learning by Context Prediction

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Revisiting Self-Supervised Visual Representation Learning

    VIEW 1 EXCERPT

    Scaling and Benchmarking Self-Supervised Visual Representation Learning

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Boosting Few-Shot Visual Learning With Self-Supervision

    VIEW 2 EXCERPTS

    Selfie: Self-supervised Pretraining for Image Embedding

    VIEW 1 EXCERPT