{
  "id": "2106.00318",
  "version": "v1",
  "published": "2021-06-01T08:48:38.000Z",
  "updated": "2021-06-01T08:48:38.000Z",
  "title": "Semi-Supervised Disparity Estimation with Deep Feature Reconstruction",
  "authors": [
    "Julia Guerrero-Viu",
    "Sergio Izquierdo",
    "Philipp Schröppel",
    "Thomas Brox"
  ],
  "comment": "Women in Computer Vision workshop CVPR 2021",
  "categories": [
    "cs.CV"
  ],
  "abstract": "Despite the success of deep learning in disparity estimation, the domain generalization gap remains an issue. We propose a semi-supervised pipeline that successfully adapts DispNet to a real-world domain by joint supervised training on labeled synthetic data and self-supervised training on unlabeled real data. Furthermore, accounting for the limitations of the widely-used photometric loss, we analyze the impact of deep feature reconstruction as a promising supervisory signal for disparity estimation.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2021-06-01T08:48:38.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "deep feature reconstruction",
      "semi-supervised disparity estimation",
      "domain generalization gap remains",
      "successfully adapts dispnet",
      "real-world domain"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}