Webly Supervised Learning of Convolutional Networks

@article{Chen2015WeblySL,
  title={Webly Supervised Learning of Convolutional Networks},
  author={Xinlei Chen and A. Gupta},
  journal={2015 IEEE International Conference on Computer Vision (ICCV)},
  year={2015},
  pages={1431-1439}
}
  • Xinlei Chen, A. Gupta
  • Published 2015
  • Computer Science
  • 2015 IEEE International Conference on Computer Vision (ICCV)
  • We present an approach to utilize large amounts of web data for learning CNNs. Specifically inspired by curriculum learning, we present a two-step approach for CNN training. First, we use easy images to train an initial visual representation. We then use this initial CNN and adapt it to harder, more realistic images by leveraging the structure of data and categories. We demonstrate that our two-stage CNN outperforms a fine-tuned CNN trained on ImageNet on Pascal VOC 2012. We also demonstrate… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    Explore Further: Topics Discussed in This Paper

    Webly-supervised learning for salient object detection
    1
    Recognition From Web Data: A Progressive Filtering Approach
    9
    WebVision Database: Visual Learning and Understanding from Web Data
    76
    Attention Transfer from Web Images for Video Recognition
    23
    Unsupervised Pre-Training of Image Features on Non-Curated Data
    40
    Harnessing noisy Web images for deep representation
    11
    NoisyArt: A Dataset for Webly-supervised Artwork Recognition
    3
    Learning without Prejudice: Avoiding Bias in Webly-Supervised Action Recognition
    2

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 75 REFERENCES
    Very Deep Convolutional Networks for Large-Scale Image Recognition
    38237
    ImageNet classification with deep convolutional neural networks
    50909
    ImageNet Classification with Deep Convolutional Neural Networks
    50909
    Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
    11160
    Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
    2595
    Weakly- and Semi-Supervised Learning of a DCNN for Semantic Image Segmentation
    368
    Harvesting Mid-level Visual Concepts from Large-Scale Internet Images
    141