{
  "id": "2004.03623",
  "version": "v1",
  "published": "2020-04-07T18:01:26.000Z",
  "updated": "2020-04-07T18:01:26.000Z",
  "title": "PatchVAE: Learning Local Latent Codes for Recognition",
  "authors": [
    "Kamal Gupta",
    "Saurabh Singh",
    "Abhinav Shrivastava"
  ],
  "comment": "To appear at CVPR 2020",
  "categories": [
    "cs.CV",
    "cs.LG"
  ],
  "abstract": "Unsupervised representation learning holds the promise of exploiting large amounts of unlabeled data to learn general representations. A promising technique for unsupervised learning is the framework of Variational Auto-encoders (VAEs). However, unsupervised representations learned by VAEs are significantly outperformed by those learned by supervised learning for recognition. Our hypothesis is that to learn useful representations for recognition the model needs to be encouraged to learn about repeating and consistent patterns in data. Drawing inspiration from the mid-level representation discovery work, we propose PatchVAE, that reasons about images at patch level. Our key contribution is a bottleneck formulation that encourages mid-level style representations in the VAE framework. Our experiments demonstrate that representations learned by our method perform much better on the recognition tasks compared to those learned by vanilla VAEs.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2020-04-07T18:01:26.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "learning local latent codes",
      "recognition",
      "encourages mid-level style representations",
      "mid-level representation discovery work",
      "unsupervised representation"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}